Lasers based on transitions of C, N, O, and Cl atoms (see Chap. 3, Sect. 3.4) have certain common features. First of all, in the active media of these lasers, consisting of the buffer gases He, Ne, and Ar, and the molecular additives CO, CO2, N2, O2, Cl2, CCl4, etc., lasing occurs at IR transitions of the atoms making up the belonging to molecules. Second, the optimal concentrations of molecular additives are very low, ~ 10~2 Torr, which is comparable with the concentration of molecular impurities in highly pure rare gases. We note that lasing was earlier observed at many transitions of the C, N, O, Cl atoms in low-pressure gas-discharge lasers [129].

Lasing mechanisms of NPLs based on C, N, O, and Cl atoms have not as yet been finally established owing to the variety of plasma processes in active media which contain a large number of neutral particles (atoms, molecules, radicals), various types of positive and negative ions, as well as to the lack of information regarding the rate constants of many of them. The processes that are suggested as populating for laser levels include dissociative excitation during collision of a metastable atom of buffer gas with molecules [130, 131], collisional-radiative recombination of atomic ions [131, 132], as well as recombination of positive and negative ions [133].

Despite the sizeable lack of information noted above, IOFAN associates pro­posed quite detailed kinetic models for NPLs based on transitions of Cl [134], C, N [135], and O [136] atoms. It was proposed that the basic channels of populating of levels in lasers operating on C, N, and O atoms are the reactions of three-body recombination C+(N+,O+) + e + M! C*(N*,O*)+ M (M is buffer gas atoms or electrons). In a laser operating on transitions of the Cl atom, along with the three — body recombination reactions, processes of dissociative recombination of molecu­lar ions ClJ and CCl^, and ion-ion recombination of an excited ion (CClJ)* with a negative ion Cl2 were suggested. Unknown constants were determined during adjustment of calculation results to experimental data.

Turbulence can lead to perceptible refractive index fluctuations. The danger of wave front distortion and noticeable coherent light beam attenuation arises. But because the effective thermal diffusivity of a turbulent flow considerably exceeds the conventional molecular thermal diffusivity that is inherent in laminar flows and stationary media, the rate of change of temperature, density, and refractive index inhomogeneities increases noticeably.

In the opinion of Schlichting, in a turbulent flow “the details of the pulsating motion superimposed on the main motion are so complex that to describe them theoretically seems hopeless” [2]. A great deal of time has passed since this statement was made. However, despite the longstanding efforts of specialists, the development of reliable mathematical models for the structure of turbulent flows that would make it possible to perform quantitative calculations with acceptable accuracy are far from complete. Nevertheless, based on concepts developed by Kolmogorov [8], certain regularities can be identified that make it possible in a number of cases to qualitatively describe the behavior of the systems under consideration and even to make approximate quantitative estimates [9, 10]. We will examine the structure of a turbulent flow using the model from ref. [8] and assess the potential effect of turbulence on the optical quality of a medium.

Unlike the schemes presented previously, which use the NOC principle, some studies [65—68] consider the possibility of creating a powerful pulsed-periodic RL with direct conversion of the kinetic energy of the fission fragments into laser radiation. It is proposed that this RL may be used in inertial confinement fusion facilities as a final amplifier for the laser driver [65, 66].

A schematic of a pulsed-periodic RL based on the separation of the reactor and subcritical laser block is shown in Fig. 11.15. The laser block is a cylindrical longitudinal cavity containing the core of the IBR-type “start-up” pulsed reactor

[64] with a reactivity modulator. It is filled with a laser medium containing fissile material and includes elements of the neutron moderator. Neutron calculations [67, 68] have shown that the system operating at a frequency <5 Hz with about 8 MJ of energy release in the laser block over one pulse can provide ~1 kW/cm3 of specific power deposition in the laser medium at a duration of 1-5 ms. Some parameters for this RL are given in Table 11.6.

Upon evaluation of the energy characteristics of the laser radiation, it was supposed that the conversion efficiency of the nuclear energy absorbed in the laser medium into the laser emission is щ = 2.0 %. It was proposed in studies [65, 66] that the laser medium could use inorganic aprotonic liquids (for example, POCl3-ZrCl4:Nd3+) with uranium dissolved in them (see the first section of this chapter). In the authors’ opinion [65, 66], the possibility of also using gas laser media in the design under consideration is not ruled out. However, in this case, at least the problem of effective energy deposition to the gas medium must be solved, and a laser medium must be found that not only has a high nI, but a long upper laser level lifetime (~1 ms).

Thanks to the high fluxes of nuclear radiation and the short pulse duration (~10 ns), when nuclear explosive devices are used it is possible to obtain high specific power depositions (q ~109 W/cm3 for gas NPLs). Laser media are pumped in this case basically by electromagnetic radiation in the X-ray and y-ray spectral range. Studies to develop NPLs using nuclear explosive devices were carried out up until 1987, when a ban was introduced on underground testing. The results of these experi­ments and possible applications of such NPLs are examined in Chap. 12.

Dependence of Ionization Processes on the Type of Charged Particles

Various types of ionizing radiation are used to act on the gas medium—uranium and transuranium fission fragments, a particles, у quanta, fast electrons and protons. In the case of у quanta, the gas ionization is provoked by fast electrons arising as a result of Compton scattering, the photoeffect, and effect of electron-positron pair formation. In the initial stage of ionization processes, primary ionization during the immediate interaction of the charged particle with the atoms is distinct from secondary ionization due to the interaction of atoms with electrons formed as a result of primary ionization.

The process of ionization of an atom may be viewed as a binary collision of an oncoming charged particle with one of the electrons of the atom’s shell [3]. Because

© Springer Science+Business Media New York 2015 103

S. P. Melnikov et al., Lasers with Nuclear Pumping,

DOI 10.1007/978-3-319-08882-2_4

Fig. 4.1 Normalized spectra of primary electrons in neon during ionization by uranium fission fragments (1) and by fast electrons with an initial energy of 1 MeV (2) [4]

of the great difference in the masses of heavy charged particles (for example, a fission fragment) and the electron, only a comparatively small percentage of the fragment energy can be transferred to the orbital electron. Thus the electron spectrum arising due to primary ionization by heavy particles is softer than the analogous spectrum formed during ionization of a gas by fast electrons [4, 5]. For example, Fig. 4.1 shows the spectra of primary electrons during the ionization of neon by uranium fission fragments and fast electrons with an initial energy of 1 MeV [4]. Although both spectra have maxima in the low-energy region, the spectrum formed by fast electrons has a significant percentage of electrons with energies up to ~10 keV. The average energy of the primary electrons formed in neon as a result of ionization by fast electrons and fission fragments is equal to 150 eV and 40 eV, respectively. In the case of fission fragments, these electrons may provide an additional one or two acts of ionization (on average), while in the case of the fast electrons, it is from five to ten [5]. Information about the average energy of primary electrons in ionization by fission fragments for other rare gases agrees with the above data for neon. It follows from the calculations in [6] that the average energies of primary electrons in the ionization of He, Ar, Kr, and Xe by fission fragments are 60 eV, 44 eV, 41 eV, and 35 eV, respectively.

Despite the specifics of primary ionization by heavy and light charged particles discussed previously, differences in the radiation effects on the gas media of different types of ionizing particles are not substantial, because the ultimate result is a combined effect of primary and secondary ionization. This is confirmed by the data cited below, from which it follows that the final electron energy distribution and the energy cost of forming an ion-electron pair in different gases are virtually independent of the type of charged particles. The same conclusion can be drawn from analysis of the luminescence spectra of gas media and the characteristics of gas NPLs, which depend not on the type of the charged particles, but on the specific power deposition. In this regard, the same kinetic models are used to calculate the characteristics of NPLs excited by heavy charged particles and those excited by fast electrons.

One of the chief parameters of NPLs is the lasing energy (power), which is unambiguously associated with the parameter e, the share of kinetic energy of fission fragments absorbed in the gas mixture with respect to their total kinetic energy released in the uranium layers. The thickness of the active layer usually is comparable with the fragment range in the material making up this layer. The methods developed by now [14—17] make it possible, for given parameters of laser cells (geometric dimensions, material of active layer, composition of gas mixture), to calculate e as a function of gas density and thickness of the active layer, which is assumed constant. Within this approach the structure of the layer is considered to be uniform. But in fact, using the presently existing methods of thin layer’s manufacturing, the above conditions (thickness invariability and uniformity) are difficult to achieve. A real layer is non-inform, and its surface is rough [24, 25]. The amplitude of the deviations in layer inhomogeneities from the average value of its thickness, and the frequency of their repetition, can depend directly on the tech­nology of deposition of the fissile material on the substrates, and on the resistance of the layers to external environment factors, such as ejection of fuel micro-particles, erosion, corrosion and the like.

Study [26] showed that the presence of inhomogeneities in the uranium layer must lead to a reduction in the resultant efficiency in comparison with a uniform layer. This reduction must occur independently of whether the roughness of the active layer surfaces exerts a screening effect on the escape of fragments into the gas or not.

Several papers [16,17] contain information on the characteristics of a stationary RL with longitudinal gas circulation relative to the axes of the laser channels. Calcu­lations for this design were made by employees of the IOFAN and MIFI.

A schematic of one RL channel (cell) is shown in Fig. 10.3. In this case, an RL is considered with a cylindrical core consisting of neutron moderators and a pene­trated system of parallel channels, which are filled with a laser gas medium. A critical mass of uranium is accumulated from the uranium layers deposited on the internal surfaces of the channels. According to one study [16], the optimal diameter of the laser channel, which is determined by the path length of the fission fragments in the gas medium and the profile of the energy deposition function along the channel radius, should be approximately 30-40 % of the path length of the fission fragments in gas. For example, the path length of fragments for helium at room temperature and 3 atm is 3.4 cm (see Table 7.1), and the optimal diameter of the channel according to [16] is 10-14 mm. At a reactor core temperature of ~1,000 K, the fragment path length and the optimal channel diameter for helium at 3 atm are equal to 11 cm and 30-45 mm, respectively.

Neutron calculations of the critical RL parameters as functions of the volume concentration of the moderators in the core, the diameter of laser channels, and the thickness of the uranium layer were performed in a diffuse multigroup approxima­tion. BeO, Be2O, and graphite were examined as neutron moderators. As shown by the calculations, each moderator has a certain minimum of critical R and H (radius and height of the cylindrical core), which is a function of the volume concentration of the moderators in the core. With the laser channel diameter at 20 mm, a core with a BeO moderator has the minimal dimensions R = H = 80 cm at an 80 % volume concentration for the moderators (the volume concentration of the gas medium, therefore, is 20 %). The base version for further examination was chosen as a core with a BeO moderator surrounded by side and end neutron reflectors made of 15-cm

Fig. 10.3 Schematic of RL channels [16]: (1) neutron moderator; (2) the metal tube of a laser channel with a layer of fuel; (3) laser gas medium; (4) channel with coolant

thick beryllium. The results of neutron calculations for this variant of the RL are given in Table 10.1.

The output performance of the base version RL using He-Xe as the laser medium is given in Table 10.2. To operate this RL in stationary mode, gas flowing must be provided along the closed path at a sufficiently high velocity ~150 m/s. This requires significant power consumption.

As the authors note [16], this variant of the RL is not definitive, because choosing the optimal design for the RL is a difficult problem with a large number of constraints for different parameters, many of which are known to a poor degree of accuracy. It should be pointed out here that at a gas velocity of U = 150 m/s, the Reynolds number, Re = U/dv ~105, significantly exceeds the critical Recr = 2,300, therefore the gas flow will be substantially turbulent. This is a serious obstacle to realizing the laser operation (see also Chap. 9, Sects. 9.1 and 9.2).

Following the successes with noble gas pumping, various laboratories undertook new NPL studies with three main goals: development of shorter wavelength lasers, lower neutron thresholds, and improved efficiencies. A NPL with visible output was developed in 1977 by a University of Illinois/Sandia team using an He-Hg charge exchange scheme [20]. Even shorter wavelengths were predicted when Miley’s team reported in 1978, for the first time, that gain had been measured with nuclear pumping of excimers [55]. Then, in the search for lower thresholds the University of Illinois group discovered a new class of NPLs where lasing occurs in a minority species (i. e., “impurity” lasers). The first such laser used small concentrations of N2 in neon [56]. Subsequently, a low-threshold (0.04 W/cm3 pump power) laser was developed using carbon from the dissociation of CO and CO2 during irradiation [57]. Later, an 3He-Ne laser with an even lower threshold was reported at the University of Florida, but the results are disputed [58, 59]. Later, gain measure­ments were reported on 3He-Ne at low fluxes or order 108 n/cm2 s [60].

VNIITF began studying NPL problems in 1979, and developed the EBR-L exper­imental complex for these purposes [2, 3, 31]. At present, apart from the EBR-L pulsed reactor, the complex includes a set of laser cells, a system of evacuation and

Fig. 2.15 Diagram of EBR-L reactor core: [3, 31]:

(1) mechanism for fine adjustment of reactivity,

(2) upper copper reflector,

(3) hemispherical parts of reactor core, (4) cylindrical copper reflector,

(5) polyethylene moderator,

(6) lower reflector,

(7) safety unit, (8) pulse rod. (9) stop unit filling of laser cells with gas mixtures, an arrangement to withdraw optical radiation from the reactor hall and register its parameters. Ten types of NPLs were studied with this complex, and lasing was achieved at 21 laser transitions [10, 55], including metal-vapor lasers (Cd, Zn, Hg) and laser at UV transitions of the molecular ion N2+.

The EBR-L reactor (Figs. 2.14 and 2.15), which is a modified version of the pulsed reactor EBR-200, was put in operation in 1981. The reactor core consists of two hemispherical parts made of 235U-Mo alloy (3 % molybdenum by mass), enclosed by a copper reflector. Between the two parts of the reactor core there is a cylindrical copper reflector with a through channel 12.5 cm in diameter. A polyethylene moderator 60 cm in long is placed in this channel, and it accommo­dates the laser cells. Some characteristics of the EBR-L reactor are shown in Table 2.6.

The average flux density of the thermal neutrons along the length of the uranium layer at the pulse maximum is 1.3 x 1017 cm~2 s_1.

A scheme of one of the laser cells for metal-vapor NPLs is shown in Fig. 2.16. The central part of the cell is in the form of a stainless steel tube 80-cm long, with a wall thickness of 3 mm. A three-section wire heater is wound around the tube and is used to heat the tube up to 850°K. Water-cooled extension pipes and adjustment units are connected to the ends of the tube. The total length of the cell is 120-

140 cm. Inside the cell there are tubes with an internal diameter of 2.8 cm, on whose surface a layer of U3O8 is deposited (90 % 235U enrichment) at a thickness of 2­3 mg/cm2 and a length of 60 cm. For experiments with NPLs, laser cells are also used without heaters, with uranium layer diameters of 1.2 and 4.8 cm. The basic characteristics of the cells are shown in Table 2.7.

To measure the characteristics of laser or luminescent radiation emerging from the cell, optical equipment (Fig. 2.17) is used, which can determine the laser spectra, energy and threshold characteristics of the laser pulse. In experiments to study the luminescent characteristics of gas and vapor media, the spectral composition of radiation, the temporal pulse shape, and the intensity of luminescence were mea­sured for several lines. The spectral range for registration of optical signals was 0.2-5 qm.

Fig. 2.17 Diagram of the withdrawal and registration of optical radiation [3, 31]:

(1) laser cell, (2) adjustment laser, (3) calorimeter,

(4) semitransparent plate,

(5) focusing lens,

(6) spectrograph,

(7) deflecting mirror,

(8) photoreceivers,

(9) monochromator

5.1 Introduction

This chapter examines the kinetics of plasmochemical processes and the lasing mechanisms of various types of gas NPLs pumped using neutrons from pulsed nuclear reactors. The lasers considered here typically have Continuous Wave,

i. e. CW operation (or, more accurately, quasi-CW) lasing mechanisms, because the characteristic times of basic plasma processes are much less than the durations of pulsed reactor pumping pulses (>50 ps). To determine the lasing mechanism, it is necessary to ascertain the processes of populating and quenching both of the upper and of the lower lasing levels. Information about the specific features and the constants of the most important plasmochemical processes is contained in a number of studies (see [1—4], for example).