`Original Russian Text © V.P. Zimakov, V.A. Kuznetsov, N.G. Solovyov, A.N. Shemyakin, A.O. Shilov, M.Yu. Yakimov, 2016, published in Fizika Plazmy, 2016, Vol. 42, No. 1,
`pp. 74–80.
`
`LOW-TEMPERATURE
`PLASMA
`
`Interaction of Near-IR Laser Radiation with Plasma
`of a Continuous Optical Discharge
`V. P. Zimakov, V. A. Kuznetsov, N. G. Solovyov, A. N. Shemyakin,
`A. O. Shilov, and M. Yu. Yakimov
`A. Ishlinsky Institute for Problems in Mechanics, Russian Academy of Sciences,
`pr. Vernadskogo 101-1, Moscow, 119526 Russia
`e-mail: yakimov@lantanlaser.ru
`Received March 19, 2015
`
`Abstract—The interaction of 1.07-μm laser radiation with plasma of a continuous optical discharge (COD) in
`xenon and argon at a pressure of p = 3–25 bar and temperature of T = 15 kK has been studied. The threshold
`power required to sustain COD is found to decrease with increasing gas pressure to Pt < 30 W in xenon at p >
`20 bar and to Pt < 350 W in argon at p > 15 bar. This effect is explained by an increase in the coefficient of
`laser radiation absorption to 20−25 cm–1 in Xe and 1−2 cm–1 in Ar due to electronic transitions between the
`broadened excited atomic levels. The COD characteristics also depend on the laser beam refraction in plasma.
`This effect can be partially compensated by a tighter focusing of the laser beam. COD is applied as a broad-
`band light source with a high spectral brightness.
`
`DOI: 10.1134/S1063780X15110100
`
`1. INTRODUCTION
`Continuous optical discharge (COD) [1] was
`obtained for the first time in 1970 [2, 3]. At present, it
`is applied as a plasma-based light source with a high
`spectral brightness in a wide wavelength range [4, 5].
`Such light sources can be used in microelectronics,
`photoemission electron microscopy, and ellipsome-
`try, as well as to measure the optical characteristics of
`materials. Practice shows that the most efficient is
`application of COD-based light sources in the far UV
`range (λ < 300 nm).
`These applications became possible due to the
`appearance of high-efficiency near-IR lasers, in par-
`ticular, diode lasers and ytterbium fiber lasers, as well
`as owing to the relatively low threshold power required
`to sustain COD by these lasers (a few tens of watts in
`xenon and several hundred watts in argon at high pres-
`sures [4–6]). When the laser power substantially
`exceeds the threshold value, the fraction of laser radi-
`ation absorbed by the COD plasma can reach 80% [4].
`This, together with the high efficiency of fiber (30%)
`and diode (50%) lasers, provides high conversion effi-
`ciency (up to 15–20%) of the input electrical power
`into the broadband plasma radiation, close to that of
`traditional plasma-based light sources, such as xenon
`arc lamps.
`The possibility of sustaining COD in high-pressure
`noble gases by lasers with a wavelength of λ ≈ 1 μm at
`such low threshold powers [4–6] was rather unex-
`pected. The matter is that, in the first experiments on
`
`sustaining light combustion waves [7, 8] performed
`with millisecond neodymium lasers (λ = 1.06 μm)
`with a high pulse energy, it was established that the
`threshold power required to sustain plasma in argon
`and air at atmospheric pressure amounts to several
`hundred kilowatts. This is about two orders of magni-
`tude higher than the threshold power for plasma sus-
`taining by СО2 lasers with λ = 10.6 μm.
`The same relation between the threshold powers
`required to sustain COD by radiation with different
`wavelengths follows from the consideration of mecha-
`nisms for the absorption of laser radiation by plasma in
`the continuous spectrum described by the Unsoeld–
`Kramers formula [1, 10]. Taking into account photo-
`ionization absorption (the Unsoeld correction), the
`contribution of which at λ = 1 μm is much higher than
`that at λ = 10.6 μm, insignificantly affects the above
`power ratio.
`Estimation of the threshold power for sustaining
`COD according to the method [1] based on the calcu-
`lation of the coefficient of laser radiation absorption
`by the Kramers formula with allowance for only the
`inverse bremsstrahlung mechanism yields about two
`orders of magnitude overestimated values of the
`threshold power as compared to those observed in [4–
`6]. Although just the Kramers formula was used by the
`authors of [5] when elaborating a theoretical model,
`they overlooked that their theory disagreed with
`experimental results, probably, owing to the applied
`character of that work.
`
`68
`
`Energetiq Ex. 2013, page 1 - IPR2015-01300, IPR2015-01303
`
`
`
`INTERACTION OF NEAR-IR LASER RADIATION
`
`69
`
`
`
`Fig. 1. Mutual arrangement ofthe COD plasma, initiating
`electrodes, and laser beam: (I) bulb with the gas and elec-
`trodes, (Z) COD plasma, (3) conventional boundary ofthe
`laser bm, and (4) lens with a focal lengthf(d is the beam
`diameter in front of the lens). The false grey scale photo
`illustrates the initiation ofCOD from an arc discharge (the
`plasma glow is taken with a I05-fold attenuation).
`
`plasma and its location near the focus. The threshold
`power level was achieved by either decreasing the laser
`power at a fixed pressure or decreasing pressure as the
`lamp was being cooled at a constant laser power. Thus,
`the pressure range in which COD was observed was
`limited from below by the maximum power ofthe laser
`and from above by the filling pressure of the lamp with
`allowance for its heating. The obtained experimental
`dependences for xenon and argon are shown in Figs. 2
`and 3. The experimental points were approximated
`according to the formula P,(p) = P,, + P,, which follows
`from the analysis ofthe energy balance with allowance
`for laser radiation absorption and the removal of the
`absorbed energy due to heat conduction and radiation
` is derived from
`3l'lA(Tk)
`the radiative loss power ofa cylindrical plasma layer of
`and P is
`l~11(Tk )
`derived from heat conduction losses, where |.r,_ ~ p2A2 is
`the coefficient of laser radiation absorption and 8 ~ p2
`is plasma emissivity. It can be seen that P, is propor-
`
`radius R0 at a temperature T,‘
`
`[1]. In this fonnula, P, =
`
`In the present work, results are reported for the first
`time from the experimental study of the influence of
`absorption mechanisms on the threshold power for
`sustaining COD in xenon and argon by laser radiation
`with a wavelength A = l.O7 pm at different gas pres-
`sures.
`
`2. SUSTAINMENT OF COD BY LASER
`
`RADIATION WITH A = 1.07 pm
`
`The experiments were performed in several stages
`with different continuous and pulse—periodic fiber
`ytterbium lasers (manufactured at the NTO IRE-
`Polus, IPG Photonics Corp.) with powers P from
`200 W to l.5 kW. The transverse distribution of laser
`radiation was close to a Gaussian beam or a mixed
`low-order mode. Some results were obtained with a
`
`YRL—QCW laser operating in the millisecond pulsed
`mode with a power from 300 W to 1.5 kW. In the
`focused beam of this laser,
`the propagation of a
`plasma wave was observed during one laser pulse from
`the instant ofdischarge initiation by an arc channel to
`the fonnation of a steady-state plasma, the energy
`balance conditions in which were close to those in a
`
`COD sustained by continuous radiation. The center
`of the lasing band corresponded to A = 1.07 pm, and
`its width was from 3 to 5 nm, depending on the radia-
`tion power.
`
`Plasma was initiated by a short-tenn arc discharge
`and then sustained by focused laser radiation inside a
`sealed quartz bulb of a standard xenon arc lamp with
`the filling pressure po from 8 to 15 bar or a refillable arc
`lamp filled with xenon or argon at a pressure p (Fig. l).
`The focusing parameter F = f/d, affecting the shape
`and stability of the COD plasma (see Section 5 and
`also [9]), determines the beam converging angle or
`within which 86% of the laser radiation power propa-
`gates. The experiments were carried out at F values
`from 3.3 to 15 and higher, at which a z l/Ff
`
`The above pressure range is of interest because it
`is the range in which the mechanism is manifested of
`laser radiation absorption due to bound—bound
`electronic transitions between energy levels of
`excited xenon atoms under conditions of strong
`Stark broadening in dense plasma that is in local
`thermodynamic equilibrium (LTE). At
`the laser
`wavelength A = 10.6 pm, bound—bound transitions,
`as well as photoionization, play a minor role; thus,
`the absorption in this case is caused only by
`free—free transitions according to the inverse brems-
`strahlung mechanism.
`
`First of all, we experimentally determined how the
`minimum (threshold) power P, required to sustain
`COD depends on the gas pressure p. The dependence
`of the threshold power on the parameter F turned out
`to be weak, because, near the threshold, refraction of
`laser radiation in the region of its interaction with
`plasma was small due to small dimensions of the
`
`PLASMA PHYSICS REPORTS Vol. 42 No. l
`
`2016
`
`Energetiq EX. 2013, page 2 - |PR2015-01300, |PR2015-01303
`
`Energetiq Ex. 2013, page 2 - IPR2015-01300, IPR2015-01303
`
`
`
`70
`
`ZIMAKOV et al.
`
` and depends weakly on p аnd that Ph ∼
`
`3. ABSORPTION OF NEAR-IR LASER
`RADIATION IN PLASMA
`Figures 2 and 3 also show theoretical dependences
`of the COD threshold power with allowance for
`absorption only in the continuous spectrum (curves 2),
`with allowance for the contribution of strongly broad-
`ened atomic absorption lines corresponding to transi-
`tions between high-lying excited energy levels of xenon
`and argon (curves 3), and with allowance for the con-
`tribution of transitions from the lowest excited 4s
`argon levels and 6s xenon levels (curves 4).
`The COD threshold power was determined from
`the analysis of the balance between laser radiation
`absorption and energy losses due to heat conduction
`and radiation. The coefficient of laser radiation
`absorption in plasma was calculated by the Kramers
`classical formula with the Unsoeld correction, which
`takes into account photoionization [10]. To take into
`account absorption caused by intra-atomic transitions
`in strongly ionized plasma at a high pressure, we
`
`20
`
`R
`
`tional to
`1/p2.
`2 Xe
` = 7.7 bar2 kW
`It was found that, for xenon,
`p P
`h
`2 Ar
`Xe
` = 26 bar2 kW
`and
`= 7 W, whereas for argon,
`rP
`p P
`h
`Ar
`and
`= 240 W. The quantity p2Ph has the meaning
`rP
`of the threshold power for sustaining COD at p ≈ 1 bar
`under conditions in which the energy balance is dom-
`inated by heat conduction, while Pr is the threshold
`power at high pressures under conditions of the radia-
`tive balance. These experimental data agree with the
`results obtained in [4–6] under other conditions and
`in narrower pressure ranges.
`In our experiments, we also measured the powers
`of laser radiation incident on the plasma and transmit-
`ted through it, the spatial distribution of the laser
`intensity, and the spectral brightness of thermal
`plasma radiation from COD.
`
`Xe, F = 310
`
`Pt, W
`104
`
`Ar, F = 8
`
`4
`
`8
`
`12
`
`p, bar
`
`1 2 3 4
`
`103
`
`102
`
`1 2 3 4
`
`10
`
`20
`
`p, bar
`
`Pt, W
`103
`
`102
`
`20
`
`Fig. 2. Threshold power Pt for sustaining COD in xenon as
`a function of the gas pressure p: (1) experimental data (dif-
`ferent points correspond to different values of the focusing
`parameter F in the range from 3 to 10); (2) theoretical esti-
`mation with allowance for both the inverse bremsstrahlung
`mechanism of laser radiation absorption and photoioniza-
`tion (the Unsoeld−Kramers formula [10]); (3) the same as
`in curve 2, plus the contribution of the processes of laser
`radiation absorption involving bound electrons of xenon
`atoms with excitation energies of E ≥ 9.4 eV; and (4) the
`same as in curve 2, plus the contribution of the processes
`involving bound electrons with E ≥ 8.3 eV.
`
`Fig. 3. Threshold power Pt for sustaining COD in argon as
`a function of the gas pressure p: (1) experimental data
`obtained with the focusing parameter F = 8; (2) theoretical
`estimation with the coefficient of laser radiation absorp-
`tion calculated by the Unsoeld−Kramers formula [10];
`(3) the same as in curve 2, plus the contribution of the
`absorption processes involving bound electrons with exci-
`tation energies of E ≥ 12.8 eV; and (4) the same as in curve
`2, plus the contribution of the processes involving bound
`electrons with E ≥ 11.6 eV.
`
`PLASMA PHYSICS REPORTS
`
` Vol. 42
`
` No. 1
`
` 2016
`
`Energetiq Ex. 2013, page 3 - IPR2015-01300, IPR2015-01303
`
`
`
`INTERACTION OF NEAR-IR LASER RADIATION
`
`71
`
`applied an approach similar to the Unsoeld approach
`describing photoionization, as it was done in [11] to take
`into account the contribution of strongly broadened
`lines to the integral plasma radiation. The minimum
`COD threshold powers observed in our experiments
`corresponded to the values of the absorption coefficient
`μ ≈ 20–25 cm–1 in Xe and μ ≈ 1–2 cm–1 in Ar.
`For xenon (Fig. 2) at relatively low pressures on the
`order of the atmospheric one, the threshold power for
`sustaining COD corresponds to the absorption mech-
`anisms described by curve 2, disregarding the possible
`contribution from intra-atomic transitions. This is
`because, at such pressures, the atomic absorption lines
`are relatively narrow and the laser wavelength λ =
`1.07 μm is far from strong absorption lines of xenon. In
`the range of middle and high pressures, the Stark
`broadening of spectral lines makes it necessary to take
`into account the contribution to absorption from high-
`lying excited states of xenon (curve 3). As the pressure
`increases further, the overlapping wings of spectral
`lines form a pseudo-continuum. In this case, it is nec-
`essary to take into account the integral contribution of
`all excited xenon states (curve 4). Thus, the theoretical
`dependences calculated with allowance for different
`components of the coefficient of laser radiation
`absorption (Fig. 2) demonstrate that the contribution
`of electronic transitions between excited xenon states
`gradually
`increases
`in with
`increasing pressure
`because of the broadening of the energy levels of the
`xenon atom.
`For argon (Fig. 3), good agreement was obtained
`without allowance for the levels lying below the 4p
`group (the excitation energy 12.8 eV), because the
`energy of a photon with the wavelength λ = 1.07 μm
`
`(hν = 1.16 eV) is lower than the energy of the 4s–4p tran-
`sition of argon even in the case of strong broadening.
`For xenon, the 6s–6p transitions lie closer to the
`laser radiation band and, due to the broadening, the
`role of 6s–6p transitions increases with increasing
`pressure, which leads to an increase in the absorption
`coefficient and the corresponding decrease in the
`COD threshold power at high pressures.
`It follows from these data that it is possible to fur-
`ther decrease the COD threshold power by decreasing
`the laser wavelength. Comparison of the structures of
`the argon and xenon terms allows us to conclude that,
`for radiation with the wavelength λ = 0.96 μm (hν =
`1.3 eV, which is close to the energy of the 4s–4p tran-
`sition in argon), the pressure dependence of the COD
`threshold power should be similar to the analogous
`dependence for radiation with the wavelength λ =
`1.07 μm in xenon with a correction for the higher ther-
`mal conductivity of argon. Thus, it can be expected
`that, for a diode laser with the wavelength λ = 0.96 μm,
`the threshold power of sustaining COD in argon at a
`pressure of p > 16 bar is at a level of Pt ≈ 100–150 W,
`i.e., about three times lower than that at the wave-
`length λ = 1.07 μm.
`
`4. THERMAL RADIATION AND PARAMETERS
`OF COD PLASMA
`The brightness of the COD plasma is determined
`by its dimensions and temperature, which are estab-
`lished as a result of the balance between laser radiation
`absorption and plasma energy losses. Figure 4 shows
`the experimental data on the spectral brightness Ip(λ)
`of the COD plasma sustained by laser radiation with
`the wavelength λ = 1.07 μm in xenon at a pressure of
`
`1234
`
`0.48
`0.36
`
`0.24
`
`400
`
`600
`
`800
`
`λ, nm
`
`Ip, W сm2 nm1 sr1
`6
`
`100
`
`0.3
`220
`
`Fig. 4. Spectral brightnesses Ip of the COD plasma and an arc discharge in xenon at a pressure of p = 22 ± 2 bar (the COD is
`sustained by laser radiation with the wavelength λ = 1.07 μm and different powers P; the focusing parameter is F = 3.3): (1) arc
`discharge, (2) COD at P = 65 W, (3) COD at P = 85 W, and (4) COD at P = 230 W. The dashed lines show the results of calcu-
`lations of the spectral brightness of a plane slab of LTE xenon plasma with a temperature of T = 15 kK and pressure of p = 22 bar
`for different slab thicknesses: s = 0.24, 0.36, and 0.48 mm. The calculation method is described in [10, 11].
`
`PLASMA PHYSICS REPORTS
`
` Vol. 42
`
` No. 1
`
` 2016
`
`Energetiq Ex. 2013, page 4 - IPR2015-01300, IPR2015-01303
`
`
`
`72
`
`ZIMAKOV et al.
`
`Xe, F = 10
`
`Fig. 5. Two different states of COD in xenon at a pressure
`of p = 20 ± 2 bar, laser power of P = 146 W (λ = 1.07 μm),
`and focusing parameter of F = 10. Plasma is shown by con-
`tour lines of the brightness plotted with a step of 10% of the
`maximum brightness against the background of a coordi-
`nate mesh with a cell size of 0.1 × 0.1 mm. Laser radiation
`is incident from left to right. The straight lines show the
`optical axis and the beam boundary in the absence of
`plasma. The state of “short” plasma is on the top, and the
`state of “long” plasma with two temperature maxima is on
`the bottom. Both states can be stable at the given parame-
`ters [9].
`
`p ≈ 22 bar. The figure also presents the comparison of
`the brightnesses of the COD plasma and the arc dis-
`charge of a high-pressure xenon lamp. Data for the
`COD and arc discharge were obtained by comparing
`the measured brightnesses with the brightness of a ref-
`erence source by using the same device under the same
`conditions. It can be seen that, as the laser power rises,
`the COD brightness increases and becomes several
`times higher than the brightness of the arc discharge,
`especially in the UV range, which indicates the higher
`temperature of the COD plasma. The high COD tem-
`perature is also indicated by the increase in the inten-
`sity of xenon ion lines as compared to the spectrum of
`the arc discharge.
`The temperature of the COD plasma can be esti-
`mated by comparing the COD spectral brightnesses
`measured at different laser powers with those calcu-
`lated by the method proposed in [10, 11] for a plane
`uniform plasma slab that is in the LTE state with a
`definite
`temperature. Since
`the experimentally
`observed increase in the COD spectral brightness with
`increasing laser power was accompanied by an
`increase in the plasma size in the observation direc-
`tion, it can be supposed that the plasma temperature
`did not change and variations in the plasma brightness
`were caused by the change in the thickness of the radi-
`ating plasma layer. It turned out that, for the tempera-
`ture of LTE plasma of T = 15 kK and the plasma slab
`thickness in the range of s = 0.24–0.48 mm, which
`corresponds to the actual transverse size of the COD
`plasma (see Fig. 5), the observed and calculated spec-
`tral brightnesses were in good agreement in a fairly
`wide spectral range (Fig. 4). These calculations are
`estimating in character, because they do not take into
`
`account variations in the temperature and plasma
`emissivity along the observation line. However, since
`the LTE approximation holds for the COD plasma,
`the calculated plasma temperature can be regarded as
`a certain average temperature, leaving aside the strong
`temperature dependence of the plasma brightness.
`The decrease in the actual plasma brightness as com-
`pared to the results of model calculations for the wave-
`length range shorter than 300 nm is due to the fact that
`the UV component is mainly radiated by the hottest
`central COD regions, which occupy a small volume
`and, thus, make a small contribution to the integral
`radiation. This is a manifestation of the inhomogene-
`ity of the COD plasma, which was disregarded in the
`calculations.
`
`5. EFFECT OF LASER RADIATION
`REFRACTION ON THE COD BEHAVIOR
`An important factor affecting the plasma charac-
`teristics is refraction of laser radiation on the density
`gradients of neutral and charged plasma particles.
`Refraction leads to defocusing of the laser beam in the
`region where it interacts with the COD plasma and,
`thus, affects the shape, dimensions, and location of
`plasma. A specific feature of refraction of radiation
`with a wavelength of λ ≈ 1 μm in the COD plasma is
`that the density gradients of neutral atoms and free
`plasma electrons introduce comparable contributions
`to the resulting refraction [9]. When the influence of
`refraction is strong, the plasma brightness increases
`weakly (or even decreases) with increasing laser power.
`The refraction effects can be compensated by a
`tighter focusing of the laser beam. The higher the pres-
`sure (and, accordingly, the higher the absorption and
`refraction of laser radiation in the gas and plasma), the
`tighter the focusing should be. For a small focusing
`parameter F (tight focusing), the COD plasma is
`located near the focus, its relative elongation is nearly
`proportional to F, the plasma absorbs well laser radia-
`tion, and its brightness increases with increasing radi-
`ation power (see Fig. 4). To sustain plasma with small
`linear dimensions and a high brightness, one should
`focus the laser radiation by a lens with a small focusing
`parameter F. If the parameter F exceeds a certain value
`that depends on the gas pressure and the relative role
`of refraction, the plasma elongates and acquires a
`structure with two or three temperature maxima along
`the laser beam axis (the state of “long” plasma in
`Fig. 5). As the parameter F increases further, stable
`operation of COD is violated, plasma shortens step-
`wise, its rear edge removes away from the focus, and
`the COD passes into a mode with low laser radiation
`absorption (“short” plasma). There is a rather wide
`intermediate region of F values in which both above
`plasma states are observed at the same values of the
`laser radiation power and gas pressure, i.e., a hysteresis
`effect takes place. In this case, the state of “long”
`plasma
`is
`implemented when
`the
`laser power
`
`PLASMA PHYSICS REPORTS
`
` Vol. 42
`
` No. 1
`
` 2016
`
`Energetiq Ex. 2013, page 5 - IPR2015-01300, IPR2015-01303
`
`
`
`INTERACTION OF NEAR-IR LASER RADIATION
`
`73
`
`increases, while the state of “short” plasma in the
`same range of parameters is implemented when the
`power decreases, or one observes random transitions
`from one state to another. Such plasma behavior is
`typical of COD sustained by near-IR laser radiation;
`for the first time, it was observed in [9], where it was
`called “COD bistability.”
`In addition to the tighter focusing, there are other
`methods of compensation for the refraction. For
`example, it is possible to reduce refraction in the
`beam−plasma interaction by affecting the plasma with
`a coaxial gas flow supplied along the laser beam axis in
`the propagation direction of laser radiation, as it was
`done in [12]. The refraction effect decreases due to the
`displacement of the plasma boundary under the action
`of the flow closer to the focus, where the curvature
`radius of the wave front of laser radiation is smaller
`and, accordingly, the relative defocusing action of the
`density gradients of the gas and plasma is also reduced.
`It should be noted that, when laser radiation is inci-
`dent on the plasma vertically (from bottom to top),
`plasma stability increases in comparison with the hor-
`izontal incidence of the beam due to the stabilizing
`effect of the convective flow generated by the plasma
`itself.
`
`6. CONCLUSIONS
`Our study has shown that the governing role in the
`absorption of laser radiation with λ ≈ 1 μm by the
`plasma of a COD operating in xenon or argon at an
`elevated pressure is played by transitions between
`excited states of atoms. It is specific features of the
`structure of the energy levels of atoms and their broad-
`ening at high pressures that provide sufficient absorp-
`tion of near-IR laser radiation by the COD plasma.
`Due to its high temperature and high pressure, the
`COD plasma is a compact bright light source in a wide
`spectral range. Due to the availability of fiber and
`diode lasers and permanent improvement of their
`characteristics, COD-based radiation sources have
`found widespread application.
`
`Refraction of laser radiation strongly affects the
`shape, location, and brightness of COD. Depending
`on the pressure and sort of gas, the refraction effect
`can be partially compensated by a tighter focusing, as
`well as by supplying a coaxial gas flow into the COD
`region.
`
`ACKNOWLEDGMENTS
`We are grateful to the NTO IRE-Polus (IPG Pho-
`tonics Corp.) for providing us with high-power fiber
`ytterbium lasers. This work was supported by the Rus-
`sian Foundation for Basic Research, project no. 13-
`08-00141.
`
`REFERENCES
`1. Yu. P. Raizer, Sov. Phys. Usp. 23, 789 (1980).
`2. Yu. P. Raizer, JETP Lett. 11, 120 (1970).
`3. N. A. Generalov, V. P. Zimakov, G. I. Kozlov, V. A. Ma-
`syukov, and Yu. P. Raizer, JETP Lett. 11, 302 (1970).
`4. D. K. Smith, US patent No. 7435982 (2008).
`5. S. Horne, D. Smith, M. Besen, M. Partlow, D. Stol-
`yarov, H. Zhu, and W. Holber, Proc. SPIE 7680,
`76800L (2010).
`6. D. Sperber, H.-A. Eckel, S. Steimer, and S. Fasoulas,
`Contrib. Plasma Phys. 52, 636 (2012).
`7. F. V. Bunkin, V. I. Konov, A. M. Prokhorov, and
`V. B. Fedorov, JETP Lett. 9, 371 (1969).
`8. I. A. Bufetov, A. M. Prokhorov, V. B. Fedorov, and
`V. K. Fomin, in Proceedings of General Physics Institute
`(Nauka, Moscow, 1988), Vol. 10, p. 3 [in Russian].
`9. V. A. Kuznetsov, N. G. Solovyov, A. N. Shemyakin,
`A. O. Shilov, M. Yu. Yakimov, and V. P. Zimakov,
`Proc. SPIE 8600, 860002 (2013).
`10. Yu. P. Raizer, Gas Discharge Physics (Nauka, Moscow,
`1987; Springer-Verlag, Berlin, 1991).
`11. G. I. Kozlov, V. A. Kuznetsov, and V. A. Masyukov,
`Sov. Phys. JETP 39, 463 (1974).
`12. V. P. Zimakov, A. Yu. Kedrov, V. A. Kuznetsov,
`A. N. Shemyakin, N. G. Solov’yov, and M. Yu. Yaki-
`mov, Proc. SPIE 7913, 791305 (2011).
`
`Translated by L. Mosina
`
`PLASMA PHYSICS REPORTS
`
` Vol. 42
`
` No. 1
`
` 2016
`
`Energetiq Ex. 2013, page 6 - IPR2015-01300, IPR2015-01303