Work on NPLs narrowed to four laboratories in the 1980s: the Sandia National Laboratory, Idaho National Environmental Laboratory (INE), University of Illinois, and University of Missouri. The general thrust focused on increasing efficiencies through improved mating with the reactor system. NPL research at INE employed the TREAT pulsed reactor and has explored several possible laser systems. How­ever, no results have been published to date. Sandia National Laboratory used the Sandia Pulsed Reactor III (SPR-III) for experiments at high excitation rates over small volumes, and the Sandia Annular Core Research Reactor (ACRR) at lower rates but larger volumes. The energy deposition rate in the laser medium with SPR-III was maximized to increase the range of lasers that could be studied [61, 62]. Laser excitation rates obtained were ~10 kW/cc with 3He excitation [63] and —2 kW/cc with fission-fragment excitation [5]. Sensitive techniques were developed at ~0.35 pm for measuring small-signal gains as low as 0.5 %/pass (using a tunable probe laser) and laser cavity losses as low as 1 %/roundtrip. With XeF, a gain coefficient of 0.7 %/cm was observed at 0.351 pm with 3He excitation at several atm [9]. Intermittent gains ~0.5 %/cm were also observed with the tunable probe laser for fission-fragment excitation of XeF gas mixtures using Ar buffer gas at ~1 atm. However, attempts at lasing with either type of excitation at ~2 kW/cc were unsuccessful [9]. Possible explanations include too low of a pumping rate, transient gas lasing, and a non-optimum gas mixture. Models of the entire fission-fragment excitation process have been developed in collaboration with University of Illinois researchers [63]. Measurements were also made of the index gradients at moderate to high pressures with uranium wall coatings [3].

More recently, studies were performed on the reactor-pumped atomic Xe laser at both 1.73 and 2.03 pm (first reported in [34]). Lasing was observed with the ACRR and SPR-III reactor facilities [37], and Rigrod experiments were performed to measure small-signal gain, saturation intensity, and non-saturable losses. Pump power densities ranged from 5 to 1,000 W/cm3. Small-signal gains such as 0.7 %/ cm were measured at 1.73 pm in Ar-Xe mixtures, with reported laser energy efficiencies ~3 % and power efficiencies ~5 %. Small-signal gains ~1-3 %/cm were measured at 2.03 pm in He-Ar-Xe mixtures, with energy efficiencies ~2.4 %. Non-saturable losses appeared to be small [64, 65]. Lasing was also observed at ~2 pm in He-Ar at the lower pump rates of the ACRR. Other Sandia experiments studied extraction of laser energy from a larger portion of the excited region through the use of unstable laser resonators and the development of specialized equipment for excitation of larger volumes of laser gas. Adaptive optic concepts have been intensely studied as to counter act “thermal blooming” as described in reference [35].

Studies at the University of Illinois pioneered the development of three new types of NPLs: (1) O2(1A)-I2; (2) nuclear-pumped flashlamp I2; and (3) 3He-Ne-H2. These are briefly described next.

Measurements by using the University of Illinois TRIGA showed relatively efficient production of 02(1Д) in low-pressure irradiations. A scaled-up design based upon these results would use ozone production at higher pressure followed by photodissociation with a nuclear-pumped flashlamp to form 02(1 Д) [66]. Direct lasing on the 02(1Д) using a “п-pulse” technique was also proposed by Scully and colleagues [67]. This is based on the concept that a pulsed laser can efficiently extract energy from an inverted molecular medium, even though the transitions involved are highly forbidden, for example singlet delta oxygen which lives for 40 min in its excited state. This extraction is accomplished by injecting pulses that are strong enough to drive the molecules from the excited to the weakly coupled ground state. A well-known situation in which this occurs is provided by Rabi flopping of the molecule from the excited to the ground state via a so-called “п-pulse.” Thus, for example, one envisions 02(1Д), NF(1 Д), or 52(1Д) molecular medium produced by collision with fission fragments in the reactor and the gas injected into an optical cavity. A “п-pulse” produced by a dye laser is then injected into the inverted medium that “sweeps” the energy from the system. Further detailed discussion is given in study [68].

The use of a nuclear flashlamp to pump a laser [49] was first demonstrated experimentally at the University of Illinois by Williams and Miley [49—51]. The nuclear-pumped flash lamp used 3He at 2,000 Torr with 0.5 Torr CHBr3 and 200 Torr Xe. This mixture was in a hollow cylinder surrounding the laser cavity, which was filled with various pressures of C3F7I. Stimulated XeBr* fluorescence pumped an atomic iodine laser at 1.31 qm. This experiment was extremely impor­tant in demonstrating the potential for nuclear flashlamp pumping, which was discussed in a number of theoretical articles. While the efficiency of the flashlamp was less than anticipated, subsequent kinetic studies suggested modifications in the mixture ratio that could potentially result in improved operation [49].

A subsequent University of Illinois discovery involved the use of H2 as a quenching agent to achieve lasing at 0.58 qm in a nuclear-pumped 3He-Ne-H2 mixture [50]. This NPL had a much lower pump threshold than reported earlier for 3He-Ne-Ar mixtures (5 versus 120 W/cm3). An efficiency of order of 1 % was anticipated. This laser then appears to be well suited for various space power beaming applications. Further details are given in the presentations [69, 70].

The RF State Science Center FEI (in Obninsk) began studying problems of NPLs in 1981 at the initiative of V. Ya Pupko [39]. Within the scope of the program to develop a powerful laser amplifier with nuclear pumping, at present an evaluation unit of the laser system has been developed which was called “Stand B” [13, 39, 56]. Here the characteristics of one of the two parts of “Stand B” are examined, the so-called “first workplace,” which was put in operation in 1996, and was intended to study the characteristics of individual laser cells. Information about the “second workplace” of Stand B is provided in Sect. 6.3 of Chap. 6.

The basic element of the laser stand is the BARS-6 dual-core pulsed reactor, made on the basis of the BARS-5 reactor [3, 32] in accordance with the design documentation and technical assistance of VNIITF. The two reactor cores are arranged on a platform that can be moved through the reactor hall depending on the tasks of the experiment. The distance between the axes of the reactor cores is varied between 340 and 1,500 mm, making it possible to irradiate laser cells more uniformly than in the case of a pulsed reactor with a single reactor core. The half­height duration of the reactor pulse is 150-200 ^s.

In one of the positions (“first workplace”) of the reactor core, experiments are carried out to study the characteristics of individual NPLs (see Fig. 2.1e). By way of example, Fig. 2.18 shows a diagram of one of the experiments studying the characteristics of an Ar-Xe laser [13]: the distance between the reactor cores is 70 cm; the cylindrical stainless steel laser cell has a length of 250 cm and internal diameter of 49 mm (wall thickness 0.5 mm). A 5-^m thick layer of 235UO2 was deposited to the internal surface. The distribution of the specific power deposition

along the length and radius of the laser cell at the reactor pulse maximum for the Ar-Xe mixture (100:1) at a pressure of 0.5 atm is shown in Fig. 2.19.

In experiments in the “first workplace” of Stand B, studies were also carried out on the gain measurements of some laser media [57, 58] and the dynamics of development of optical non-uniformities [59].

The active media of gas NPLs are mostly binary mixtures A-B, where A is a buffer gas with high ionization and excitation potential, and B is a lasing additive with lower ionization and excitation potential, the concentration of which is 1-10 %. In this case, the energy deposited in the buffer gas A is transported virtually without losses to the lasing additive B at the expense of the processes of charge-transfer, the Penning processes, and transfer of excitation, while the quenching of excited atoms B* by their own atoms is insignificant [5].

Populating of the upper laser levels of the atoms and ions of B can occur at various stages of the sequence of relaxation processes (see Fig. 4.16). The first stage involves the processes of energy transfer from ions and atoms of the buffer gas A to atoms of the additive B:

© Springer Science+Business Media New York 2015 143

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

DOI 10.1007/978-3-319-08882-2_5

1. The charge-transfer process A+(Aj)+B! (B+)*+A(2A). Depending on the pressure of the buffer gas, atomic ions A+ or molecular ions Aj will take part in this reaction.

2. The Penning reaction A*+B! (B+)*+ A + e. The reaction takes place in the case when the energy of the excited atom A* is more than the ionization potential of the atom B.

3. The process of excitation transfer A* + B! B* + A.

The upper lasing levels can be populated by means of any of these three processes. For example, the first two are fundamental in populating the lasing levels of He-Cd and He-Zn NPLs that radiate on transitions of the excited ions Cd+ and Zn+ (see Chap. 3, Sect. 4.4), while in an Ar-N2 UV laser excited by electron and ion beams [6­8], the levels C3Пи of the molecule N2 are populated as a result of the last process.

The processes of recombination play an important role among plasma processes; the main ones are three-body recombination of atomic ions A+(B+) + 2e(e, A)! A* (B*) + e(A) and dissociative recombination of molecular ions Aj(Bj, AB+) + e! A *(B *)+ A(B, A).

Lasing levels can also be populated as a result of recombination processes. For example, as a result of three-body recombination, inverse population may occur between the excited states of the ion Sr+ (A = 430.5 nm) [9]. When atmospheric pressure lasers are pumped by nuclear radiation and the specific power deposition is comparatively low (q < 5 x 103 W/cm3, concentration of electrons the processes of dissociative recombination of molecular ions Bj or AB+ with electrons whose rate constants for heavy rare gases are sufficiently high— 10—7 — 10—6 cm3/s [10]—are of interest. As a result of dissociative recombination, one of the atoms is formed in an excited state; the number of such states is not great, which assures selectivity of populating of the upper lasing levels. First considered in study [11], such a populating mechanism occurs in NPLs operating on transitions of the atoms Ar, Kr, Xe, Ne, and Hg.

A high rate of depopulating (or quenching) of lower lasing levels is necessary for efficient operation of lasers in CW mode. In CW gas-discharge low-pressure lasers, as a rule the depopulating of these levels is owed to radiative decay. In NPLs operating with the gas medium at atmospheric pressures, such depopulating occurs basically through inelastic collisions: with atoms of the medium (NPLs operating on transitions of Ar, Kr, and Xe atoms), with electrons of the plasma (certain metal vapor NPLs), and as a result of Penning reactions when an atom in a lower lasing state collides with atoms or molecules of an additional additive (NPLs operating on transitions 3p — 3s of the Ne atom). In study [12], which contains the results of investigation of A-NF3 lasers (A = He, Ne, Ar) during excitation by an electron beam and transverse discharge, harpoon reactions A* + NF3 ! AF* + NF2 having high rate constant values up to~ 10—9 cm3/s were used for depopulating of lower levels. We note that the idea of using inelastic collisions for selective depopulating of excited states of atoms was put forward by V. A. Fabrikant as far back as 1940 [13, 14].

The mechanisms for forming inverse population of levels in the main types of NPLs reviewed here are briefly described in Table 5.1.

As noted in Chap. 4, when gases are irradiated with nuclear particles, a plasma is formed with the recombination type of non-equilibrium, which leads to preferred populating of excited states “top-down,” owing to the processes of recombination and other relaxation processes. Therefore, for NPLs it is advisable to introduce the ultimate efficiency (nmax) as the ratio of the energy of laser quantum hv to the energy expenditures for formation of one active particle (ion or excited atom of buffer gas) [5]:

Птах = (1 + f * /f+)hv/wi = (f++ f *)hv/q, (5Л)

where wi is the energy cost of formation of an ion-electron pair (see Chap. 4, Sect. 4.2); f+, f* are the formation rates of ions and excited atoms when the medium is irradiated by nuclear particles (for rare gases f*f+ = 0.53 [15]).

Formula (5.1) was written for the case when both ions and excited atoms of the buffer gas take part in the formation of upper laser levels. If only ions of the buffer gas take part in populating the upper laser levels, the formula for qmax has a simpler form:

Птах = hv/wi = f + hv=q. (5-2)

In contrast to the quantum efficiency of gas discharge lasers, the ultimate efficiency for NPLs does not depend on the location of the upper laser level relative to the ground state.

The calculation results of ultimate efficiencies from formula (5.1) for a number of active media are given in Table 5.2. When some of these mixtures (He-Xe, Ar-Xe, He-Ne-M, He-Cd, He-Zn) were used, lasing was observed during nuclear pumping, while for other mixtures lasing was obtained in the afterglow of the gas discharge [16]. The value of qmax grows with an increase in the energy of the laser photon or with a reduction in wt. Naturally, the real efficiency of the nuclear — pumped laser will depend not only on qmax but also on many other parameters:

the correlation of rates of relaxation plasma processes, the selectivity of the populating process of the upper laser level, and the rate of depopulation of the lower laser level. Thus, the choice of active media with a high value of щтах is a necessary, but not sufficient, condition for creation of efficient NPLs.

In the final stage of the relaxation processes, metastable states Bm are formed in the A-B mixture, which are destroyed as a result of formation of the excimer molecules B2* and subsequent radiative decay. However, under some conditions, the process of associative ionization Bm + Bm! B+ + B + e can become the basic channel of destruction of the states Bm. In this case, atomic ions B+ form again, and can again participate in populating the upper laser levels, and consequently the ultimate efficiency will be greater. According to estimates [17], for the He-Ar mixture, the ultimate efficiency as a result of this recirculation may be about twice as high.

Sequential population of excited states “top-down” in a nuclear-excited plasma makes it possible to organize a cascade of laser transitions. In this case, it becomes possible to obtain simultaneous lasing in different spectral ranges and to increase the efficiency. One can propose several lasing cascade schemes [5]: (a) a cascade of laser transitions between excited levels of an atom (ion) when the lower laser level of one transitions is simultaneously the upper level of the transition below; (b) sequential lasing: first operating on transitions of the ion, and then (after recombination of the ion), operating on transitions of the atom; (c) when using multicomponent mixtures, sequential lasing on transitions of atoms (ions) of indi­vidual components.

As an example, Fig. 5.1 shows one possible scheme of cascade lasing. When the ternary mixture He-Ne-Xe is used, simultaneous lasing can occur in the visible spectral range at the transition 3p-3s of the Ne atom (A = 585.3 nm) and in the IR range at the transitions 5d-6p of the Хе atom (A = 1.73, 2.65, or 2.03 ^m). In this

case, xenon is used to depopulate the lower lasing 3 s-level of the Ne atom and is simultaneously a laser component. But if we take the additive M, which efficiently quenches the long-lived 6s state of the Xe atom in collisions, then a cascade of laser transitions is possible between the levels 5d-6p and 6p-6s of the Хе atom. We note that in study [18], when the ternary mixture He-Ar-Xe is excited by an electron beam, one version of cascade lasing is obtained: simultaneous lasing at transitions of the Ar atom (A = 1.79 ^m) and Xe atom (A = 2.03 ^m).

Dynamics of Near-Wall Passive Zone Development

It was previously shown that a so-called passive zone is formed in direct proximity to the side wall of a laser cell with deposited uranium layer, from which heat is intensely removed to the wall. Large positive density and refractive index gradients up to Vn ~ 10~4 cm-1 characterize a passive zone. Light rays that have a direction

a yA,

Fig. 8.12 Dependence of active lasing region and passive zone interface variation upon time in a cylindrical cell (1) and an infinitely extended plane-parallel cell (2): (a) a cell filled with a mixture based on He (Р0 = 1.5 atm); (b) a cell filled with a mixture based on Ar (P0 = 0.35 atm) parallel to the cell axis at any point in a passive zone are deflected toward its walls (at least, after several cavity passes). Consequently, this zone should not make a contribution to a laser beam’s output energy. Passive zone size, l, increases over time. During the initial stage, when l << d (d is the transverse dimension of the cell), passive zone size variation in time is adequately communicated by depen­dence (8.3). Outside the confines of this zone (in the midsection of the gas volume) the medium density and refractive index gradients are negative, during which I Vn I ~ 10~6 cm-1. Lasing develops exactly in this very region.

At l ~ d, the dependence of near-wall passive zone size noticeably deviates from the regularity expressed by (8.3). This zone’s development is most demonstrably observed based on the example of a cell irradiated by a time-constant neutron flux. Calculation results for the dependence of active region and passive zone interface variation upon time are presented in Fig. 8.12. Two types of cells were examined: a cylindrical cell with an inside diameter of d = 2 cm and an infinitely extended planar cell, in which the distance between the uranium layers was also d= 2 cm. In Fig. 8.12, the distance in the direction of the y axis is reckoned from the axis of symmetry for the cylindrical cell and from the plane of symmetry for the planar cell. As in the experiments described in [47], the transverse dimension of the aluminum outer cell walls to which metallic uranium layers with a thickness of 5U = 5 mg/cm2 were deposited equaled 0.5 cm. In both cases, the layers were coated with a protective aluminum film that had a thickness of SA = 0.5 mg/cm2. These are the same layers with a protective coating that were used during the experiments described in [47]. Calculations were performed for an Ar-Xe mixture (70:1) at a pressure of P0 = 0.35 atm and an initial temperature of T0 = 293 K (Fig. 8.12а), as well as for a He-Xe mixture (2000:1) at P0 = 1.5 atm and the same temperature (Fig. 8.12b). In these cases, the cell’s normalized transverse dimension equaled D0 = 0.26. The results presented in Fig. 8.12 were obtained under the assumption that, from a moment in time of t = 0, neutron flux density remains constant

Ф = 3.5 x 1014 cm_2-s_1. The results show that without gas flowing at irradiation times exceeding 41 ms for Ar and 38 ms for He in the cylindrical cell, or 74 ms (Ar) and 47 ms (He) in the infinitely extended planar cell, a heat removal zone characterized by a positive refractive index gradient spreads over the entire gas mixture volume. From a moment in time of t ~15 ms, a steeper approach of the active region boundary, yA(t), to the center is observed as compared to correlation (8.3). In addition to the specific role of gas-dynamic effects, this is explained by the nonlinear dependence of the gas heat conductivity coefficient upon temperature [16]. The effect of the geometric shape of the cells is also obvious.

The first studies of the luminescent properties of inorganic liquids doped with europium and neodymium and excited with uranium fission fragments were appar­ently conducted at VNIIEF [3]. Experiments on the VIR-2 pulsed reactor with uranium fission fragment (specific energy deposition up to 25 J/cm3) excitation of a POCl3-SnCl4-235UO22+:Nd3+(Eu3+) solution with a uranyl concentration of ~2 x 1019 cm~3 obtained low conversion efficiencies nX < 0.1 % [3, 18]. The conversion efficiency for perchlorates with europium under analogous conditions was significantly lower.

The spectral-luminescent characteristics of laser media based on inorganic liquids with excitation by nuclear particles under various conditions were studied in more detail in studies at the FEI. The main results of these studies are given in

reviews [19, 20]. The following inorganic liquids were examined as test objects: POCl3-SnCl4:Nd3+(Eu3+,Tb3+) [21—27]; ZnCl2-GaCl3:Nd3+(Tb3+, Mn2+) [22]; SOCl2-GaCl3 :Nd3+ [22, 25]; SO2Cl2-GaCl3:Nd3+ [25, 28]; and D2O:Eu3+(Tb3+) [22, 29], doped with Nd3+, Eu3+, Tb3+, Mn2+.

The main effort was directed toward the search for laser-active liquid media containing 235U. For this purpose, technology was developed to introduce uranium into laser media in the form of uranyl UO22+ in concentrations <1020 cm-3, and the interactions between uranyl and ions of rare-earth elements were studied [30, 31]. The laser media were excited with the products of nuclear reactions taking place upon the interaction between the nuclei and neutrons. The neutron sources were either the accelerator KG-0.3 with a neutron energy of 14 MeV [22, 23, 25, 28, 29] operating in pulsed and continuous modes or the BARS-6 pulsed reactor [26, 27]. In the first case, during continuous operation of the accelerator, the specific pump power of the laser media was ~5 x 10~7 W/cm3 (for liquids based on D2O, it is approximately five times higher), and in the second case, while using a medium containing uranium, the specific energy deposition reached 30 J/cm3 at a pulse duration of 140 ^s.

The studies of liquid laser media for NPLs carried out at FEI obtained informa­tion on the luminescence spectra, conversion efficiency, and lifetimes of the upper laser level of media based on inorganic liquids as a function of activator concen­trations, and also their characteristics were compared using different methods and excitation modes. Some experimental information on the conversion efficiencies and lifetimes of upper laser levels for different media is given in Fig. 11.3 and Table 11.3.

The data cited testify that the conversion efficiency depends on the type of solvent and activator, the concentration of the activator, and also the presence of uranyl in the solution. The maximum value, цл = 1.4 %, was obtained for the solution SOCl2-GaCl3:Nd3+. However, with the addition of the uranyl UO22+ to

Fig. 11.3 The dependence of the conversion efficiency (Пі) on the concentration of Nd3+ for the inorganic liquid POCl3-SnCl4-O2 + :

Nd3+ in the absence of uranyl (filled circle), and also with a uranyl concentration of 1 x 1019 (open circle), 2 x 1019 (open square), 3 x 1019 (open triangle) and 5 x 1019 (open diamond) cm-3 [21, 22]

this solution, processes begin that lead to uranyl transition from a 6-valent state to a 4-valent state, making the laser characteristics of the medium drastically worse [20, 25]. Therefore, for subsequent experiments, the laser liquid POCl3-SnCl4- UO22+:Nd3+ was selected and studied in the most detail.

The possibility for reaching the lasing threshold when pumping laser media with nuclear radiation not only depends on the spectral and luminescent characteristics of the medium, but also on its radiation resistance, that is, the ability to retain optical uniformity and transparency at the lasing wavelength under the radiation influence. As was already noted, the main causes that could lead to the appearance of detrimental losses in a liquid laser medium excited with heavy charged particles are the formation of gas-vapor bubbles (light scattering) [9, 32] and radiolysis on the tracks of the fission fragments, and also thermal optical disturbances (thermal lens and optical wedge) caused by non-uniformity of the energy deposition over the active laser volume [18].

Apparently, these specific circumstances were the reasons for the absence of laser action during attempts to pump the liquid laser medium with uranium fission fragments in experiments on the BARS-6 pulsed reactor [20, 33]. Figure 11.4 shows a schematic of experiments, which used the laser liquid POCl3-SnCl4-235UO22+: Nd3+ with equal neodymium and uranyl concentrations of 3 x 1020 cm-3. A 34-cm long liquid-filled laser cell surrounded by a polyethylene high-speed fast neutron moderator was placed alongside one of the BASRS-6 reactor cores or between two cores (Fig. 11.4 shows the second variant). The specific energy deposition to the laser medium over a 140 ^s pulse is 32 J/cm3. The laser cavity was formed by two mirrors with reflectivities of 100 and 65 %.

One area of NPL development is the search for active media containing molecular gases as the laser component. In the active media of these lasers, consisting of mixtures of rare gases He, Ne, and Ar with molecular gases CO, CO2, N2, O2, Cl2, CCl4, F2, NF3, etc., lasing can occur at the transitions of the C, N, O, Cl, and F atoms. Attempts to achieve lasing at the transitions of these atoms under nuclear pumping began immediately after development of the first NPLs. In study [129], on excitation of the mixture 3He-Ne-O2 (2,000:30:1) at a pressure of 0.8 atm with nuclear reaction products 3He(n, p)3H, the observed small-signal gain was ~10-4 cm-1 at the transition of the O atom (A = 844.6 nm). These experiments were carried out on a TRIGA reactor. Somewhat later, this reactor was used to conduct experiments [130] in which, when the mixture Ne-N2 (Р < 0.5 atm) was pumped with uranium fission fragments and nuclear reaction products 10B(n, a)7Li, lasing occurred at transitions of the N atom with A = 862.9 and 939.3 nm. The output power was 1.5 mW, n ~ 10 .

Pumping of a laser at the transition 3p1P1-3s1P10 of the C atom (A = 1.45 qm) with nuclear radiation (in this case with nuclear reaction products 10B(n, a)7Li) was carried out for the first time in experiments [131, 132] on a TRIGA reactor at pressures of the mixtures He(Ne)-CO and He(Ne)-CO2 ranging from 0.03 to 1 atm and partial pressures of CO (CO2) of up to 0.02 Torr. The output power was ~1 mW (nl ~10-3 %). The laser threshold was reached at ФгА = 3 x 1014 cm-2 s-1. Later the analogous experiments were performed in France on the Caliban-pulsed reactor [133].

There is no detailed information on research of NPLs outside Russia using the transition of the Cl atom. The reviews [40,126] reported that lasing was observed at the 1.59 qm line at the mixture 3He-Cl2 (Р = 0.8 atm). The laser threshold was reached at ФгА = 7 x 1015 cm-2 s-1, and the output power was ~0.1 W.

In Russia, experimental research of NPLs operating on transitions of the atoms C, N, O, and Cl were conducted at VNIIEF. In the first experiments with the VIR-2 reactor, carried out in 1978, lasing occurred at transitions of the C atom (A = 833.5; 940.6 and 1,454 nm) while the mixtures He(Ne, Ar)-CO were excited by uranium fission fragments, and the maximal output power was 7 W (A = 1.45 |im) [134].

In subsequent experiments [135—139] on the VIR-2M reactor, a cycle of inves­tigations were carried out to find lasing at transitions of the atoms C, N, O, and Cl and to optimize laser characteristics with regard to the pressure and composition of mixtures and the cavity parameters. The LUNA-2M setup was used in these experiments. The basic results of the investigations are shown in Table 3.10. We note that earlier, lasing in nearly all of the wavelengths cited in Table 3.10 was obtained in lower-pressure gas discharge lasers [18].

The energy parameters obtained at VNIIEF for lasers operating on transitions of C, N, O, and Cl atoms are two or three orders of magnitude greater than the parameters of analogous lasers studied in other laboratories, while the laser using

the transition of the Cl atom (A = 1.59 ^m) is one of the most low-threshold of the NPLs. A specific feature of C, N, O, and Cl lasers is the low concentration (~10-2 Torr) of laser component (molecular gas), compared to the concentration of molecular impurities in high-purity rare gases. Therefore, to increase the energy characteristics it is necessary to have a higher degree of exhaustion of the laser cells, and additional purification of the initial gases.

The LUNA-2M setup was also used to perform experiments to find lasing at oxygen atom lines 777.4 and 844.6 nm (mixture of He-O2 and He-Ne-O2) [139] and fluorine atom lines in a range of 620-910 nm (mixture He-NF3) [140]. The exper­iments yielded no positive results, although these oxygen lines and a number of lines of the fluorine atom showed high intensity in luminescence research [81, 140, 141]. The absence of lasing at fluorine atom lines 703.8 and 712.8 nm in experi­ments with the mixture 3He-NF3 using the APRF reactor is also reported in the review [126]. We note that with electron beam pumping, lasing was obtained in the

mixture He-NF3 at the fluorine atom lines 634.8, 641.4, 712.8, 731.1, 739.8, and 755.2 nm [142].

Some studies examined lasing mechanisms and calculated the characteristics for active media that until recently did not lead to the making of NPLs, for example, active media using transitions of the 1st negative system of the CO+ molecule (A = 219 nm) [143, 144]. From calculations of the characteristics of a CO+ laser it follows that in the mixture He-CO-H2, (hydrogen is used to depopulate the lower laser levels) for q = 1-2 kW/cm3, it is possible to obtain quasi-CW lasing at A = 219 nm with rather high efficiencies: n = 3-4 % [143]; 0.7 % [144]. It should be noted that in the experiments of [145] on pumping a He-CO mixture at atmo­spheric pressure with an electron beam, stimulated radiation was registered at transitions of the CO+ molecule with A = 247 nm (1st negative system) and A = 395 and 421 nm (Baldet-Johnson system).

Authors of model [144] discussed the possibility of appearance lasing on the line of the O2 ion with A = 160 nm during excitation of the He-O2-H2 mixture by fission fragments and an electron beam. We could not establish the affiliation of this line with an electronic-vibrational transition of the O2 ion (see [146], for example). As noted in study [147], as a result of the charge-transfer process He^ + O2, there is selective populating of the state b4Xg~ of the O2 ion, from which the transitions of the first negative system of oxygen (500-850 nm) begin.

The reason for density fluctuations in a turbulent flow may be not only the gasdynamic instability of the flow, but also liquid or gas mixing in the presence of the average temperature value’s dependence upon coordinates. So, as a result of mixing in a turbulent flow, individual portions of the medium that are characterized by a certain temperature value appear at points where medium particles with different temperature values were heretofore present.

Obukhov [12] and Yaglom [13] examined the structure of a temperature field in such a flow. According to the model they developed, the linear scale of the minimum temperature inhomogeneities is determined by the correlation

le ~ (a3/w)1/4, (9-16)

where a is the thermal diffusivity coefficient.

Because the kinematic viscosity coefficient, v, and thermal diffusivity, a, for gases differ very little from one another (a Prandtl number of Pr ~ 1), the minimum linear scales of inhomogeneities, as follows from Eqs. (9.6) and (9.16), are in virtual agreement:

le/l0 ~(a/v)i/A ~ 1- (9-17)

The mechanism of temperature equalization in sufficiently large volumes of a turbulent flow is due to the combined effects of the medium’s turbulent motion and heat conduction [12]. Temperature equalization in large-scale inhomogeneities and intermediate-scale inhomogeneities primarily occurs due to the internal turbulent mixing of the smaller-scale gas particles. The typical time of this equalization is determined by the correlation

те ~ lt/u

and corresponds to the lifetime of dynamic lt-scale inhomogeneities according to the Kolmogorov model. For maximum-scale inhomogeneities

твт ~ Л/U — (9-18)

Temperature equalization in the smallest-scale inhomogeneities occurs by way of conventional molecular heat conduction [12]. The typical time required for this equalization can be estimated using the formula

ТЄо ~ le/a-

From the latter correlation and Eq. (9.14), with allowance for Eq. (9.17), we obtain

Т0О/т0 — ku0/a. (9-19)

Because Pr ~ 1 for gases (i. e., a ~ v) then from Eq. (9.19) in concordance with Eq. (9.5), as should be expected, we get те0~ т0.

Thus, the conclusion can be reached that the space-time structure of the turbu­lent flow temperature inhomogeneities originating as a result of mixing corresponds to the space-time structure of turbulent pulses with a dynamic nature.

The first experiment on the pumping of a laser by y-radiation from a nuclear explosion was presumably performed at VNIIEF in 1971. Liquid xenon emitted as Хе2* (X = 172 nm) excimer molecules were studied as an active media. This laser was pumped earlier with an electron beam by the staff at FIAN (Lebedev Physics Institute of the Academy of Science) [14].

Experiments on pumping two types of lasers with the y-radiation from a nuclear explosion were carried out in the United States in 1973. This was done in a testing area in Nevada by the staff of LLNL [15] and LANL [16].

The LLNL experiments studied xenon gas at 6.25-18.75 atm and specific energy depositions of 4.5-15.3 J/cm3. At xenon pressures higher than 9.4 atm, the gain of the spontaneous emission was recorded in the UV spectral range near 170 nm. In this case, the gain increased with the rise in pressure from 0.1 to 0.29 cm-1. The light radiation had an obviously pronounced directivity, which is additional proof that the stimulated intense radiation was present in these experiments.

The staff at LANL checked the possibility of initiating a chemical HF-laser (X = 2.7-3.0 qm) on SF6-C2H6(8:1) mixture at 1.7 atm with the y-radiation from a nuclear explosion. A 44 cm long laser cell 2.9 cm in diameter was exposed to a collimated y-radiation for 12 ns duration. The exposed length of the cell was 10 cm, and the active volume was 65 cm3. Lasing was reported in these experiments, and the laser pulse energy was 3.9 J (n = 6.5 %) at a 17 ns half-amplitude pulse duration.

Further studies of high-power gas lasers excited by the y-radiation from a nuclear explosion were carried out at VNIIEF in 1977-1984. The laser setups examined next are single-pass, traveling-wave amplifiers.

Of the powerful neutron sources, the most widespread are the nuclear reactors, which were used to perform the bulk of investigations to find active NPL media and to study their characteristics. Nuclear reactors are sources not only of neutron radiation, but also of у radiation; however, for pumping NPLs, neutrons are used, since in this case it is possible to obtain a specific energy deposition to the laser medium approximately two orders of magnitude greater than the energy deposition from у radiation.

In experiments investigating NPLs with help of reactors, as a rule direct pumping of active media is carried out using not neutron radiation, but the products of exothermal nuclear reactions, which take place during interaction of neutrons with the nuclei of 235U, 10B and 3He (Table 1.4).

For effective excitation of the gas medium, it is necessary for the isotopes that interact with the neutrons to be in immediate contact with the laser medium. When nuclear reactors are used as neutron sources, two basic types of laser-medium excitation are utilized (Fig. 1.1): (1) a gaseous isotope or compound thereof (3He, 235UF6) is a component part of the laser medium; (2) the internal surface of the gas-filled laser cell is coated with a thin layer of isotope (10B, 235U) or compound thereof (235UO2, 235U3O8). In studies published up to this time, both methods are used. From the comparison made in the first section of Chap. 7 of the efficiency of pumping gas NPLs using the isotopes He, U, and B, it follows that approx­imately identical energy contributions to gas media can be obtained using the

gaseous isotope 3He and a thin layer of 235U. However, in the first case, it is necessary to use only 3He as the buffer gas; this substantially limits the possibilities of gas mixture selection. When thin uranium layers are used, the maximal specific power deposition (q ~5 x 103 W/cm3 for gas media) is achieved in experiments with pulsed reactors with a minimal pulse duration of ~100 ^s. This pumping method is interesting because on its basis, it is possible to create powerful nuclear-laser units (reactor-lasers) in the core of which uranium layers are used not only to excite the laser medium, but also as nuclear fuel.

The specific power deposition and uniformity of excitation of laser media depend on the magnitude and duration of the neutron flux, the pumping method, the geometry and dimensions of the laser cell, the type of nuclear particles, and the gas pressure.

In the case of a volumetric source of pumping using 3He, the non-uniformity of pumping comes from the absorption of slow neutrons in 3He and from the reduction of the energy contribution in the region near the wall owing to the removal of reaction products (1H, 3H) to the walls of the cell. Results of computation of the total energy deposition and spatial distribution of the deposited energy depending on the 3He pressure and the diameter of the cylindrical cell are given in [67, 68], while [69] shows the results of computation of the energy deposition for a 235UF6- He mixture. The use of the gaseous (at comparatively low temperatures) compound 235UF6 makes it possible in principle to obtain an efficiency of nuclear energy deposition in the gas mixture of up to 100 %, but the search for nuclear-pumped gas media based on 235UF6 to date has not yielded a positive result owing to the high rates of “quenching” of excited atoms by molecules of UF6.

In the case of a surface pumping source, the area of uniform excitation is determined by the transverse dimension of the laser cell and the path length of fission fragments or a particles, which for various gases at atmospheric pressure is 1-10 cm. Computations of the total energy deposition and its spatial distribution for laser cells in the shape of a cylinder and a rectangular parallelepiped with layers of 235U applied to the internal surface, depending on the thickness of the uranium

a

Fig. 1.1 Methods of exciting NPL gas media: (a) volumetric method of pumping, (b) surface method of pumping layer, dimensions of the laser cell, and the gas pressure (as well as experimental methods of determining the energy deposition and the results of its measurement) are provided in Chap. 7. Here we note that when uranium layers are used, the efficiency of energy deposition from fission fragments in the gas medium does not exceed 50 % (an infinitely thin layer) and as a rule is 15-20 % when a thickness of uranium layer equal to one half of fragment path length in layer material (~5 mg/

cm2).

As was noted above, basically nuclear reactors were used as neutron sources in experiments with NPLs. Among nuclear reactors, pulsed aperiodic reactors [70, 71] with a pulse duration ranging from ~50 qs to ~10 ms possess the maximal possible neutron fluxes; they provide repeated and controlled fission bursts of uranium nuclei. The organization of experiments with NPLs using pulsed reactors is con­sidered in Chap. 2.

Apart from aperiodic pulsed reactors, for the pumping of NPLs it is also possible to use periodic pulsed reactors [72], although the neutron fluxes which they supply are roughly 10-50 times less than in the case of aperiodic reactors. The review [28] mentions an experiment conducted in 1985 by associates of MIFI and IOFAN using the pulsed periodic reactor IBR-30 (Joint Institute of Nuclear Research (OIYal, Dubna)). This experiment, carried out using gas mixtures He-Ne-Ar and He-Ar-Xe, did not yield a positive result, which may be explained, the authors believe, by the low power deposition.

Stationary nuclear reactors have substantially lower neutron fluxes than pulsed reactors. In stationary research reactors such as the IRT-2000 and the VVR, the thermal-neutron flux densities are ~1013 cm-2 s-1 (specific power deposition of gas media up to ~1 W/cm3), which is not sufficient for researching most NPLs, especially in the stage of the search for new laser media. Therefore, experiments with stationary reactors were directed chiefly at studying the spectral-luminescent plasma characteristics and electroionization lasers (see, for example, [54]). It should be noted that there are special SM-type stationary reactors [73], in which thermal-neutron flux densities in the central cavity of the reactor core reach

2 x 1015 cm-2 s-1. However, experiments with NPLs with an SM reactor are hampered owing to the limited volume of space with such a high neutron flux.

Of the other possible laboratory neutron sources for NPL pumping, it was proposed that the neutron radiation from tokomaks [74] could be used. Devices based on high-density plasma (plasma focus, Z-pinch) [75] could also be used for this purpose.

The options for pumping NPLs using neutron sources that were mentioned previously use nuclear reactions that take place with the interaction of nuclei of certain isotopes with thermal neutrons. To increase the flux of thermal neutrons, laser cells were surrounded by a fast-neutron moderator (Plexiglas, polyethylene, graphite). The maximal pressures of gas media do not exceed 5-6 atm, which is due to the appearance of excessive non-uniformity of pumping with a growth of pressure because of the shortening of the path length of the nuclear reaction products or weakening of the thermal neutron flux. One method of uniform pumping of gas media (at pressures of tens and hundreds of atmospheres) and of condensed media is the use of the elastic scattering of slow neutrons on atoms (nuclei) of the medium. In this case, ionization and excitation of the gas medium are carried out by recoil nuclei. This NPL pumping method was first used at VNIIEF [76] and is considered in the third section of Chap. 3 of this book.