13.1 Introduction

This book primarily discusses NPL research in Russia. The Russian program has been larger and remains active whereas work in the United States and elsewhere has dwindled. Shortly after the end of the Cold War, two circumstances combined to cause stoppage of NPL work. While the military need ceased, public opposition to nuclear power mounted following the Three Mile Island Nuclear reactor accident in the United States. Such concerns clearly carried over to a reactor pumped laser, i. e. to NPLs. Here we briefly review earlier work is the United States, prior to a build-up of effort as part of the “Star Wars” program as cold war tension grew. For completeness, the earlier history of NPL research in the United States is briefly reviewed here, largely adapted from earlier review articles [1]. Additionally, some content also appears in another previous review of NPLs [2] while some new material has been added, particularly concerning theory.

NPLs are driven by the reaction products of fission, fusion, or radioisotopes, providing the direct conversion of nuclear energy to directed optical energy. Aside from scientific aspects, research on NPLs was initially motivated because it has the potential to introduce many exciting new applications for fission reactors. This is especially true for systems where a nuclear energy source is a logical choice. The large volume made possible by using neutron excitation leads to ultra-large, steady — state power or intense pulsed-energy outputs that cannot be matched by other laser systems. Because nuclear energy is directly converted to excited states and subse­quently to lasing, a relatively high efficiency (for traditional lasers) is achieved. Furthermore, the integrated reactor-laser system can be compact because no inter­mediate conversion equipment is needed. Consequently, researchers interested in military type laser began to take note of NPLs and a classified research program was mounted.

Chapter 1 of this book reviews work in the United States quite well from the Russian point of view, and this chapter provides a brief review from the American

© Springer Science+Business Media New York 2015 S. P. Melnikov et al., Lasers with Nuclear Pumping, DOI 10.1007/978-3-319-08882-2_13

perspective as I (George H. Miley) experienced it. Due to limitations on length, this chapter cannot do justice to the major effort done in the reactors at Sandia National Laboratory and earlier work at the Idaho National Laboratory [3—9]. These projects originated by Victor George at Lawrence Livermore National Laboratory under “Star Wars” funding and were classified (interested readers may now be able to find declassified versions of some reports). It appears that the project was aimed at KW-level demonstration results when funding stopped at the end of the Cold War. Thus, this chapter might be viewed as one from the perspective of academia. Admittedly, it is also weighted toward my work at the University of Illinois that my students, and former colleagues were involved in.

Reflections on early work at Illinois are briefly noted here to bring out some of the thinking and problems faced by early NPL researchers. In addition to my early work, other early U. S. pioneers in this field included Lloyd Herwig, David McArthur, Richard Schneider, Philip J. Ebert, Herbert Helmick, and Joe Verdeyen. Indeed, follow-on work by researchers noted in this chapter has been in a sense equally pioneering in this new unexplored field.

My experiments were uniquely enabled by access to the pulsed TRIGA reactor at the University of Illinois. These studies were enhanced by special facilities at the Illinois TRIGA reactor—a graphite thermal column, a horizontal “through-port,” and a vertical access to the center core region. All three were used to great advantage at various times. Further, because I had earned a TRIGA reactor operator’s license from the NRC, we could schedule laser experiments at conve­nient times, often at nights when the lab was relatively free from complications presented when other experiments were running. Also at these times the reactor experimental bay could be darkened to aid visual observations of the optics involved.

My interest in the field of NPLs began when I attended a faculty summer school at the Idaho National Reactor site in 1963. This frequently involved a one-hour bus ride out in the desert to the historic Fast Breeder Reactor Experiment. To pass the time, I read a book about the invention of the laser, which was discovered only a few years earlier, which inspired the thought, “Why not pump the laser with nuclear energy instead of electrical?” I then mastered the physics of laser threshold calcu­lations from the book, and decided that the CO2 laser with its long wavelength in the IR, hence low threshold, would be ideal for nuclear pumping. When I returned to the University of Illinois that fall, I discussed this with Prof. Joe Verdeyen, a well — known laser researcher in the Gaseous Electronics lab at the University of Illinois. Verdeyen had already done some pioneering work on the Transversely Excited Atmospheric Pressure (TEA) CO2 Laser, and was extremely knowledgeable (he later went on to write a classic textbook on lasers, Laser Electronics, now in its third edition). He agreed with my assessment that CO2 could easily be pumped in the TRIGA reactor, even with the low neutron fluxes available with steady state operation giving neutron fluxes of ~1014 n/cm2 s. Pulsing would be “overkill.” A boron coating on the laser tube would generate alpha particles from neutron-alpha reactions, and the charged alphas entering the CO2 gas would deposit the energy needed for pumping the laser. Thus I teamed up with Verdeyen, along with several

students, and began experiments. Unfortunately, this led to several years of frustration.

Though I initially thought I had invented the idea on that Idaho bus, I soon discovered that Lloyd Herwig at Northrop Grumman in California had already proposed the concept and had actually done experiments. It is hard to be first in this age of fast-moving technology! More importantly, the University of Illinois research team slowly learned two physics phenomena that represented significant challenges to the research. First, the nuclear radiations dissociated CO2 molecules and other molecules involved in the lasing reaction. Because that effect was not included in my threshold calculations, it threw everything off. (Indeed, this has prevented nuclear pumping of CO2 to this day. Amazingly, David MacArthur did succeed with CO, and that was the first experimental NPL in the United States.)

Next Illinois researchers discovered a new problem. The alpha heating of the gas gave a radial temperature profile that peaked near the wall, causing a thermal lensing effect that defocused the internal beams and threw the cavity optics off. Indeed, when using pulses, the effect was dynamic, such that the beam deflection was large and varied with time during the pulse. The effect was vividly illustrated when a cylindrical tube open to the air but with a boron coating on the tube’s interior wall was used one night. A reference He-Ne laser beam was set up to pass through the boron coated tube located in the through beam port and projected on the building wall about 30 feet away. During the reactor pulse, the beam underwent huge spiral-circular gyrations on the wall, ultimately settling back to its original steady-state position. The effect of the gas lenses was thus abundantly obvious. The question was what to do about it to avoid this effect from spoiling the gain in the NPL cavity? Verdeyen proposed an over-focused cavity design (focal spots for internal beams were located at a point behind the cavity mirrors). That seemed to overcome the problem, although not without sacrificing some gain. (This effect was later studied in some detail by Maria Petra in her thesis work at Illinois, as discussed later.)

With this new knowledge my team continued trying to obtain lasing with CO2, in desperation going to yet higher pulse sizes. Still, all attempts failed due to the dissociation effect already noted. We did achieve and publish papers on a radiation- enhanced, electrically pumped TEA CO2 laser, i. e., this electrically driven laser output was increased by pulsing a version using a boron-coated tube. The effect could be attributed to the increased electrical conductivity created by the auxiliary ionization from the alpha particles. However, this was not the “pure” NPL we desired. So we next turned to study a possible He-Ne laser. The logic was that this medium would be less prone to dissociation interference, and again approximate threshold calculations predicted lasing could be achieved with TRIGA reactor pulses, maybe even with steady state operation. However, after over a year of frustrating results, it became evident that once again the threshold calculations were off, and achieving threshold was exceedingly challenging.

The reader may think that progress with NPL research was very slow. In contrast, research elsewhere on electrically pumped laser was making great strides. To understand the differences it must be realized that reactor-pumped laser experiments are quite different from normal lab bench laser research typical of electrically pumped lasers. The NPL is remotely located inside a tube deep within the reactor, making alignment adjustments difficult. Only a limited number of pulses could be obtained per day, and the reactor experiment schedule might only provide a limited number of days a month. Laser cavity components become activated, preventing immediate access. Also, radiation can damage optical compo­nents and detectors. Indeed, shielding is required around the detector array and in addition the laser beam must enter into the detector array without a straight path that radiation could follow. Thus, reflectors are needed to form a zigzag route. After a steep learning curve, my research group (and others elsewhere working in this field) became adept at overcoming these many obstacles. While their knowledge and technique had grown significantly, they had not yet achieved a NPL when the CO NPL was achieved by David MacArthur at Sandia National Laboratory (discussed later in this chapter). Despite this frustration, my group returned to work, even more determined to find new types of NPLs. The CO achievement was truly monumental, but due to the cooling requirements, this did not appear to be a practical result. New types of electrically pumped lasers were being reported quite frequently in those days, so why should not a variety of NPLs be found? This turned out to be the case, and, as seen from the following discussion, the University of Illinois was prolific in discovering new types of NPLs.

In experiments with pulsed and stationary reactors, optical fibers are used to withdraw the laser or luminescent radiation from the region close to the reactor core to the registering instruments located behind the biological shielding. The influence of reactor radiation on the properties of the optical fibers was examined in a number of studies (see for example [40, 45]).

Resolution of the problem of the fibers’ radiation resistance is closely associated with development of radiation-resistant optical materials. For example, develop­ment of the method for manufacturing quartz with an extremely low content of impurities made it possible to produce fibers that can be used to withdraw light in the visible range at doses of ~107 Gy [46, 47], which corresponds to the annual radiation exposure inside the core of a stationary reactor. To increase radiation resistance of fibers, small controlled impurities can be added, for example, molec­ular hydrogen [47, 48].

This section provides a brief survey of research on the luminescent spectra of gas media at high pressure in the visible and near IR spectral ranges with excitation by nuclear radiation and fast electron beams. The results of spectral-luminescence investigations in the UV range, aimed at creation of powerful excimer sources of coherent and spontaneous radiation, are not examined here, but are contained, for example, in the surveys [74, 75].

Research on luminescence provides information about the spectroscopic char­acteristics of transitions, making it possible to ascertain the kinetics of populating and “quenching” of excited states and to estimate the possibility of obtaining laser action at individual transitions. Early stages of investigations of radioluminescent characteristics of condensed and gas media performed before 1964 are considered in the survey [76].

From various methods used in spectral-luminescent investigations, we single out the method of single-photon spectrometry [39, 40, 77, 78], which is based on the counting of single photons and allows analysis of luminescent radiation depending on wavelengths and de-excitation times. Single-photon spectrometry is the basis of the method of subthreshold diagnostics of active NPL media, developed at FEI [79, 80]. Further, we shall briefly examine the most significant studies dedicated to researching the luminescence characteristics of gas media excited by ionizing radiation.

8.1 Inhomogeneities in Cylindrical Cells

The short ranges of fission fragments causes a specific energy deposition inhomo­geneities which, along with excited gas heat exchange with laser cell walls, leads to spatial density redistribution, i. e., to the violation of active medium optical homo­geneity. As noted in paper [1], with a reference to the 1980 report of A. N. Sizov and Yu. N. Deryugin, numerical and theoretical investigations of the dynamics of optical inhomogeneity origination and development in nuclear-pumped lasers first began at the VNIIEF. During first stage, these investigations only pertained to the phenomena occurring in sealed lasers excited by pulsed neutron fluxes. This important aspect of NPLs is discussed in this chapter.

The potential for coherent addition of laser channels is both interesting and impor­tant. Coherent addition methods known from the literature are based on the injec­tion of the radiation of one channel into another and subsequent separation of a specific field distribution using various selection methods.

For sets of lasers with periodic arrangements of elements, there is an opportunity for coherent phasing due to the effect of the self-reproduction of the periodic structure of the monochromatic field at some distance (Talbot optical coupling) [33]. Individual lasers may also be coupled on a common nonlinear cell in the laser cavity (mirror based on stimulated Brillouin scattering) [34].

Hereafter, as we examine a large number of optically coupled lasers, we will assume that, during independent lasing, each laser operates in a single transverse mode. If the dimensions of the active medium of an individual laser grow, the selection of a specific transverse mode turns into a problem. It can be solved using unstable cavities. Thus, the synchronization of lasers having individual independent cavities presents difficulties.

From a technical point of view, it is convenient to create sets of lasers with periodically arranged elements. Optical coupling in such sets can be implemented using external mirrors deflecting a small portion of the radiation into neighboring lasers. It is sufficient to simply use a flat mirror as the coupling element which is placed at some distance from the aperture of the set of lasers. Radiation reflected from the mirror leads to optical coupling between elements due to diffraction “spreading.” However, the efficiency of this coupling remains rather low. To increase the portion of radiation coming from one laser to another, the mirror must be moved aside. Due to diffraction, the radiation spreads along the whole aperture, and losses increase in the coupling channel. With a periodic arrangement of the set of lasers, this consideration is incorrect, because it does not take into account the reproduction effect of the periodic structure of the monochromatic field at a specific distance (Talbot diffraction coupling).

If the mirror is placed at half the reproduction distance of the periodic structure (Fig. 10.8), the image of the face plane of the channel, if it is emitted in phase, is projected onto the very same face plane. At the same time, because there is a contribution from the radiation of many channels to the image of each spot, coupling between individual lasers is sufficiently high. Without in-phase radiation of individual channels, the diffraction image breaks up, and losses increase sharply. Thus, this effect leads to selection of a phase-synchronization mode. It has been shown in study [33] that the distance z at which the mirror should be placed is equal to half of the “Talbot” distance ZT = ра2Д, where a is the lattice pitch and p is a numerical coefficient that is a function of the geometry of the lattice.

As exemplified by a multi-beam waveguide CO2 laser with periodically arranged elements, there was experimental confirmation [35, 36] of the potential for effective phase synchronization by Talbot diffraction coupling. It was shown that the graph of the laser power as a function of z can be divided into three parts. The first part relates

to independent lasing. In this case, the field distribution in the far-field region is determined by the aperture of a single channel, and the laser power is at a maximum. The second part represents a region with a partially coherent mode, with spots corresponding to interference orders are visible in the far-field and the total power drops by approximately twofold. In the third part, the radiation of each channel is highly coherent. However, the power is ~20 % the power of independent lasing.

A method that uses a spatial intra-cavity filter (Fig. 10.9) does not add up to optical coupling between lasers. The concept of a spatial filter consists of the selection of a specific configuration of the generated field that has minimal losses in the filter. If this configuration corresponds to an in-phase mode, it can be selected [36]. It was experimentally confirmed that with the use of an intra-cavity filter, there is absolutely no incoherent lasing, and complete synchronization is possible even with rough positioning of the filter. The total laser power in the principal maximum is 12 % with respect to the power of independent lasing. To enhance synchroniza­tion, it is recommended that the lengths of the laser cavities be increased and varied between themselves.

The first experiments [37] studying the coherent addition of NPL channels were performed by VNIITF researchers on reactor EBR-L. These experiments used the spatial filter method to realize an in-phase operation mode for a NPL package consisting of three parallel 60-cm long tubes with an internal diameter of 11 mm. The parameters of the laser medium, including the laser wavelength, are not given in study [37]. Diaphragms 1-2 mm in diameter were used as the spatial filters. The experiments showed that the use of a laser cavity including a telescope with a spatial filter reduces the laser radiation divergence due to phase synchronization of individual channels.

Fig. 10.10 Optical coupling in a nonlinear cell

Individual lasers may also be coupled by redistributing the beams on a common nonlinear cell in the laser cavity. SBS mirror can be used as this cell The nonlinear cell based on stimulated Brillouin scattering (SBS). This method is especially useful for NPLs that make it possible to obtain the significant laser power necessary to reach the SBS threshold. The schematic of a laser with SBS mirror is simple (Fig. 10.10). The laser radiation is somehow focused in the cell with a transparent medium. When a certain threshold power is exceeded, the radiation due to SBS is reflected back and returns to the laser cavity. If the radiation of two or more lasers is focused on one nonlinear cell, they will be synchronized due to redistribution of the energy.

To maintain a steady SBS process, the threshold radiation power Wthr, which corresponds to the threshold energy Ethr« Wthr x т (where т is the pulse duration) and is inversely proportional to the gain increment of the medium g, should enter the nonlinear medium. The energy released in the medium leads to the development of processes that compete with SBS. Therefore, when exceeding some critical energy Ecr, the effectiveness of SBS falls sharply. Thus, the determining criterion for the choice of a SBS medium is that the critical energy Ecr exceeds the threshold energy, that is, the product gEcr, which only depends on the properties of the medium. Values for the gain increment and critical energy of different media are given in Table 10.6 [34]. It follows from the table that high-purity xenon is preferred as the SBS medium.

The experiments [34] performed by VNIIEF researchers studied the effect of SBS within the range 0.69-4.2 qm. They demonstrated the potential to:

• Compensate for aberrations and increase the radiation brightness.

• Phase the radiation of a large number of laser channels (up to 1,000).

• Obtain high-frequency (>1 MHz) pulsed-periodic lasing.

In pulsed power NPLs, the pulse length is 10~3-1 s. This significantly increases the time for pressure equalization in the focal region. This means that the efficiency of SBS may be influenced by a change in refraction coefficient due to electrostriction and heating of the medium. Experiments on xenon (at 46 atm) with a wavelength X = 1.315 pm gave values determined by electrostriction: Wcr = 3.2 x 105 W and Ecr = 5 J.

An experimentally obtained a laser beam divergence for a single laser channel of ~0.5 x 10~3 rad, close to that of a diffraction one. Linear methods (serial and parallel addition, telescoping, etc.) make it possible to obtain laser beam diver­gences of up to ~10 5 rad. Use of coherent addition methods brings the beam divergence up to values no worse than 10~6 rad.