The Los Alamos Scientific Laboratory lens design program, in which a lens moves steadily toward diffraction-limited performance, samples lens performance with bundles of precisely traced skew rays, analyzes performance by calculating the image-spot sizes and positions, and optimizes performance in a least squares system that drives the lateral ray deviations toward zero. Minimizing the rms image-spot size minimizes the rms optical path difference (OPD). Minimizing the rms OPD also optimizes the diffraction modulation transfer function (DMTF). Minimizing the image-spot size and position errors minimizes and balances all seven Seidel aberrations because all seven cause lateral deviations of the rays from their ideal image points.

A general thin lens analogy between a holographic optical element (HOE) and a conventional optical element (COE) is discussed. The third order aberration theory of a HOE is given. The generalized aplanatic condition of the HOE is derived and shows that the bending of the HOE makes the aplanatic condition possible.

Waxicon mirror components offer many advantages in designing optical systems for transporting high power laser beams. The widespread use of waxicons has been limited because of their high sensitivity to tilt errors. This paper gives the equations of the surfaces of a waxicon that is rigorously corrected for both spherical aberration and coma. Computer ray tracing has confirmed its low sensitivity to tilt errors: if the aplanatic waxicon as a whole is tilted by a small angle δ, the RMS wavefront error in the output beam will be proportional to δ2.

In 1978 the author described a laser beam expander that was unobscured, corrected for spherical aberration, coma, and astigmatism, and consisted of four spherical mirrors. It was not corrected for Petzval curvature, however; therefore the expanded laser beam would defocus slightly if the input beam were not aligned to the system correctly. This paper describes a new design that has just three mirrors, one an aspheric, and that is corrected for spherical aberration, coma, astigmatism, and Petzval curvature. It is also unobscured, and has the very nice feature that the one aspheric mirror is used in a centered fashion: no off-axis section of an aspheric mirror needs to be made. The new design can be realized in any beam expansion ratio that is desired, but performance suffers at very high magnifications. Two representative designs, 4X expansion and 8X expansion, will be shown and discussed, and their performance numbers will be given.

On looking for a simplified method for dealing with the paraxial lens formulas one finds that by using the lens focal length F as a unit of measure the paraxial equations quickly simplify to a pair of graphs readily depicting the locus of all possible object and image parameters (location and size) for converging lenses (lenses with thick centers). A "thin lens" is converging if its center is thicker than its edges. A "thick" converging lens has object and image distances measured from principle planes rather than lens center as in a "thin" lens.

A plane grating predisperser has been constructed which acts as an "order-sorter" for a large plane-grating spectrograph. This combination can photograph relatively wide regions of spectra in a single exposure with no loss of resolution.

Spectral measurements often require higher quality data than that provided by available spectrophotometers. The capabilities of the digital data acquisition and analysis system for infrared spectrophotometry are presented. The system includes a Data General Corporation NOVA 2/10 computer interfaced to a modified Perkin-Elmer Model 180 Spectrophotometer. Increased accuracy is obtained through computer control of the monochromator, live zero correction at each wavelength through sample beam shutter control, and enhanced signal-to-noise ratio through multiple data sampling at each wavelength. Routine use of the system includes spectral characteristic determination of bandpass interference filters, dichroics (including polarization effect), neutral density filters, and gas samples used for calibration purposes. Spectra obtained as digital data are stored on disk for subsequent analysis including normalization, smoothing for signal enhancement, multiple spectra averaging to determine a scanning filter's effective response and extraction of such parameters as pass-band location and shape, location of minima and maxima, and spectral throughput over a given spectral range. Optical system spectral transfer functions are obtained by convolution of the spectral transfer functions determined for each optical element in a system.

Fast IR wavefront aberration sensors are essential to the improved performance of HEL systems. However, sensor designs which function well (in principle and in practice) for the incoherent blackbody realm must not be transferred over into the coherent environment without first considering possible interference problems. A case history of one such sensor which failed this precept is considered.

This paper describes a method of illuminating the region at and near the surface of a transparent sample or of a film on a transparent substrate. The features then observed, using ordinary microscopic magnification, are a speckle pattern due to surface roughness and discrete points of light due to localized point scattering sites both on and just within the surface. While the technique is similar to dark-field microscopy, here the region just within the surface is highly illuminated, and additional information concerning the size and position of scattering sites can be learned. In this technique the surface is illuminated from within the sample with a well-collimated polarized laser beam at an angle of incidence equal to or greater than the critical angle. When using an s-polarized beam, a standing wave pattern is set up just within the surface of the sample. Since the incident beam is totally reflected, the antinodes have four times the incident beam intensity and the nodes have zero intensity. This standing wave pattern may be translated and the nodal spacing may be changed by changing the angle of incidence and/or changing the laser wave-length. This allows selective illumination of various surface features. Additionally, by the use of a film of matching fluid on the surface under inspection, the roughness-caused speckle pattern may be eliminated leaving only the light coming from point scattering sites. We will describe methods of using this technique to determine certain size and positional features of the discrete scattering sites. Examples of various surfaces as viewed by this technique and other standard techniques are compared. This inspection technique has been used to study surface cleaning methods, laser damage nucleation sites, and ion milling of optical surfaces.

The Burch scatterplate interferometer is an extremely precise, equal path interferometer that provides direct wavefront information without the use of any precision optical components. This paper describes the use of Fourier transform techniques to provide a straightforward scalar diffraction theory of the interferometer. The results show the characteristic hot spot and speckle pattern seen in scatterplate interferograms. Practical applications of the theory are also discussed.

Measurements of the angular distribution of scattered light produced by specially-polished optical components are reported. The components were six 12-cm aperture coronagraph objective mirrors and two 20-cm coronagraph objective lenses. For each test the component was illuminated with white light and the magnitude of the scattered flux measured at several field points. It was not feasible to separate surface and body scattering of the lenses, so that these measurements represent the combined contributions. Successive recoatings and tests of the mirrors has led to the conclusion that the residual roughness of the substrate polish is the dominant parameter that determines the magnitude of the scattered flux. Dark ground and phase contrast measurements qualitatively support this conclusion. The fractional contributions of integrated scattered light of the lenses and mirrors are found to be of the same order, although the functional relationships are significantly different over the angular range measured.

A direct derivative measurement technique for determining Brewster's angle has been developed using a lock-in amplifier which synchronously detects the reflected intensity of a directionally modulated or dithered laser beam. In our apparatus the angle of incidence of a polarized laser beam is slowly varied while it is dithered at a high frequency by a galvanometer-mounted mirror. The modulated signal, which is proportional to the derivative of the reflectance versus angle curve, is detected with a lock-in amplifier tuned to the dither frequency. The angle at which the signal crosses zero is Brewster's angle. The angle is more precisely defined using this method as opposed to the more conventional method of determining the angle of minimum reflectance because it is easier to locate a zero crossing than a minimum in a curve. The technique has been tested at visible and infrared HeNe laser wavelengths. The precision of the measurement is typically better than five parts per thousand for materials with refractive index near 1.5 and ten parts per thousand for materials with refractive index near 2.5. Data have been gathered for a number of visible and infrared transmitting materials.

Fabrication of high quality optical surfaces requires both sensitive methods of evaluating surface characteristics and delicate control of fabrication techniques. Interferometers which yield half wave data have been augmented by computer controlled phase measuring techniques which are capable of greater accuracy and flexible real time data analysis. Key elements of the Phase Measuring Interferometer (PMI) system are a modified Twyman Green interferometer, a piezo driven reference mirror, detector, interface, and minicomputer system containing a graphic display, keyboard, and dry copier. Wavefront data are obtained by a direct measurement of phase difference over the aperture on a 32 x 32 element array. Several programs have been written to augment the capability of the basic system. System capabilities include standard isometric contour maps at variable sensitivity, wavefront subtraction and calibration modes, wavefront averaging, calculation and display of the MTF and Point Spread Function, determination of tilt and power over subapertures with respect to a defined standard aperture, and deconvolution of wavefront data to yield aberration coefficients.

A process for removing antireflection, mirror and polarizer coatings has been developed at ILC, based on work begun by LLL (Applied Optics Vol. 17, No. 12, 15 June 1978 - "Notes on Optical Coating Removal", N.J. Brown). Because of the danger (personnel hazard) involved in the hydrofluoric acid process, we employed an ammonium bifluoride solution, combined with various polishing components. The substrates, generally BK7, are fairly soft and also sensitive to chemical action. Therefore we have limited our polishing materials to aluminum oxide powder graded at 0.1 pm or smaller. For some coatings, no polishing material is used, as the ammonium bifluoride solution is adequate to remove the coating. The resulting clean surface is washed and neutralized, and is then ready for recoating.

For non-normal incidence angles, it has been shown that the polarization state may be controlled by reflection through the use of specially designed multilayer coatings. An optimization I has been used to determine layer thickness for a coating design tha. produces a 90° phase shift between the p and s polarization comppnents over a wavelength range of Δλ/λ0 = ± 5%. A particular 20 layer design produces 90° ± 1° phase shift over this spectral range while maintaining an average reflectivity of over 99.9%. A tolerance analysis on this design indicates that the coating layers must be deposited within ± 1% in order to achieve ± 3° phase shift over the spectral range.

Multilayer quarterwave stacks were fabricated to reflect at λ = 16 μm. Due to the substantial metric thickness of the stack, the mechanical stress caused the coating to separate from the substrate. The optical thickness of the lead fluoride was 35% thicker than the zinc selenide in the stress-compensated design. The tensive stress of the former compensated the compressive stress of the latter.

The I3 Sensor is a modified Hartmann wavefront sensor which combines the simplicity of the basic test with modulation techniques. The result is a sensor with vastly improved stability and dynamic range over the basic Hartmann centroid measuring technique, while retaining the high light collection efficiency of the basic system. The 13 Sensor performance has been demonstrated to approach the theoretical performance limit of wavefront phase accuracy as a function of signal-to-noise ratio. These results were obtained in a practical engineering configuration. The basic form of the sensor, modulation and refer ence techniques will be described. The performance will be compared with the basic Hartmann test and with other wavefront sensors.

Infrared interferometric systems using a CO2 laser operating at a wavelength of 10.6 μm are described. Specific systems discussed are a Twyman-Green interferometer and a Twyman-Green interferometer with an infrared computer-generated hologram (IRCGH). The reduced sensitivity due to the longer wavelength enables us to test optical components necessary for IP high-energy laser systems. This paper also illustrates typical results obtained from testing infrared trans litting materials, diamond-turned metal mirrors, and aspherics. Special alignment techniques and basic limitations of infrared interferometry are discussed.

The surface finish of a single-point diamond-turned (SPOT) mirror produced in 1977 on the large (2-meter swing) Excello lathe at the Union Carbide Corporation's Y-12 Plant in Oak Ridge, Tennessee, was such that it was not possible to do visible alignment using it. Therefore, it was necessary to polish the mirror by hand. This was done at the University of Arizona. The intent was to remove the high ridges in the SPDT mirror but not to degrade the figure. Both interferometric and encircled energy measurements were made on the mirror before and after polishing. The mirror was an f/2 off-axis parabola 39.37 cm in diameter. The equation describing the generator of the mother parabola is y2 = 309x(cm2) and the center of each off-axis sister mirror is at x = 7.64 cm, and y = 48.59 cm. After polishing it was possible to align it using techniques which employed visible light. Furthermore, the polished mirror was about 20% better as far as the rms surface figure was concerned, although cosmetically the surface finish appeared visibly degraded after polish.

A review is given of the technology and special facilities applied in the production of 18-inch diameter Polytran NaCl windows. These optical elements are to be employed in the LASL Antares CO2 Laser System. Procedures utilized in achieving window specifications are discussed.

A research program to improve large NaC1 crystal growth for CO2 laser window production is described. The program entails growth by the Czochralski technique, permitting control of crystal orientation and crystallographic integrity and relaxing the demand on purity of starting materials. A facility for producing crystals up to 20 inch diameter and 200 Lb. mass has been developed. Several 18 inch diameter windows have been produced from these after hot-forging to a fine-grain polycrystalline structure for strength. The crystals contain scattered, faceted microbubbles that convert to donut-shaped inclusions on forging. The program continues toward perfecting the new approach.

To facilitate the production of 18" diameter NaC1 windows for the Antares Inertial Confinement Fusion Program of LASL, Harshaw has an optical evaluation test facility in the same room as the polishing machines. The principal optical test instrument is an 18" Zygo Fizeau interferometer. The photographed interferogram from the interferometer is digitized manually using a BITPAD* coordinate digitizer and PET 200l** computer. The computer has been programmed (in Basic) to fit, by least squares, the fringe coordinates to the best spherical fit over the clear aperture. The output gives the best fit (sphere) and wedge (linear) and the RMS deviation of wave front from best fit. An interferogram can be reduced in less than ten minutes. The cost of the reduction equipment was less than $2,000.00.

The Helios and Antares CO2 fusion laser systems incorporate numerous large sodium chloride windows. These must be refinished periodically, making necessary a consistent and predictable polishing capability. A continuous polisher (or annular lap) which might fulfill this requirement was purchased and a process development program was undertaken at Kirtland's Developmental Optical Facility. Large NaC1 windows had not been polished on this type of machine. The machine has proven itself capable of producing λ/16 figures at 633 nm (HeNe) with extremely smooth surfaces on glass. Since then, we have been working exclusively on NaCl optics. Due to different polishing parameters between NaC1 and glass, and the slight solubility of the pitch in the slurry, this phase presents new problems. The work on glass will be reviewed. Results on NaCl to date will be reported. The potential of this type of machine relative to prisms, thin and irregularly shaped optics will be discussed.

The American Society for Testing and Materials, through the Optically Pumped Laser Section - Committee F1.02, is developing standard test methods and practices for optical testing, evaluation, and fabrication. These standards are developed within a framework of producer, user, and general interest persons by obtaining a consensus. Each test method is subjected to round-robin testing and carries with it a validated precision statement. This paper describes these optical test methods and invites suggestions and contributions of effort in the Section toward additional optical and laser standards. Nine ASTM standards are reprinted for easy reference by optical scientists.

The optical design and analysis of the LASL carbon dioxide laser fusion systems required the use of techniques that are quite different from the currently used method in conventional optical design problems. The necessity for this is explored and the method that has been successfully used at Los Alamos to understand these systems is discussed with examples. This method involves characterization of the various optical components in their mounts by a Zernike polynomial set and using fast Fourier transform techniques to propagate the beam, taking diffraction and other nonlinear effects that occur in these types of systems into account. The various programs used for analysis are briefly discussed.

The performance of the Gemini two beam carbon dioxide laser fusion system was recently upgraded by installation of optical components with improved quality in the final amplifier. A theoretical analysis was conducted in conlunction with measurements of the new performance. The analysis and experimental procedures, and results obtained are reported and compared. Good agreement was found which was within the uncertainties of the analysis and the inaccuracies of the experiments. The focal spot Strehl ratio was between 0.24 and 0.3 for both beams.

A system to determine the relative time of arrival of intense, subnanosecond, CO2 laser pulses on a fusion target is described. It is now installed on the Helios CO2 laser system, a 10 kJ, 20 TW, eight-beam CO2 facility. This beam simultaneity system is inexpensive, easy to use, and has a resolution of less than ± 5 ps.

A multi-component gaseous saturable absorber has been developed and modified to prevent the onset of parasitic oscillations coupling laser-fusion targets and the power amplifiers in the Helios CO2 laser system. This absorber mix greatly increases the extractable short-pulse energy of the laser system. A computer model that predicts the behavior of multi-component saturable absorbers has been tested against the measured performance of two different absorber gas mixes used in the Helios system. The successes and shortcomings of the model are discussed, and large and small-signal transmission data for the saturable absorber mixes and their constituent gases (used as input to the model) are presented.

This paper describes the alignment system for Helios, an eight-beam, CO2 laser which is now being used in studies involving compression and heating of deuterium-tritium filled glass spheres, 200 to 300 microns in diameter. These studies are directed towards determining the feasibility of laser-initiated fusion for commercial power generation. The laser system reached design-point output of 10 kilojoules at power in excess of 20 terawatts in June 1978.

The Los Alamos Scientific Laboratory Helios laser fusion system focuses eight powerful CO2 laser beams onto a tiny (typical 300-μm-diam) DT-filled target. The focusing is accomplished inside a vacuum chamber that is 3.5 m in diameter by 3.5 m high. The target positioning system places the target to within 5.0 μm of a predetermined point in space. This is accomplished by using two orthogonal autocollimating telescopes to determine the center of a precisely located surrogate sphere. The target is then placed at the common focal point of the two telescopes. The Helios target positioning system has been meeting or exceeding its design requirements for about one year with minimal maintenance. It has proven to be a very effective system.

Verification of satisfactory operation of the HELIOS eight-beam laser system requires measurement of many parameters of each beam on each shot. Fifty-joule samples of each of the eight 1250-J, subnanosecond 34-cm-diameter beams of the HELIOS system are diverted to a gallery of eight folded telescopes and beamsplit to provide diagnostic measurements. Total pulse energy, and prepulse and postlase energy of each beam are measured; pulse shape details and a wavelength spectrum of a selected beam from each shot are measured; and provision is made for retropulse measurement and optical quality monitoring. All data are recorded digitally in a local screen room, with control and communication through a fiberoptic link to the main HELIOS computer.

The Antares CO2 laser system is a 200-TW, 100-kJ, gas laser for the investigation of inertial confinement fusion. The light energy produced is transmitted in 72 beams extending from the input sections of the power amplifiers through the target system optics and finally focused on the target. The 72 independently-controlled beams originate from 6 power amplifiers, each having an annular array of 12 quasi-trapezoidal apertured sectors. Each of the 72 beam lines must be independently aligned and the beams independently focused on the target. The beam alignment system requirements and the physical and operational constraints are detailed and explained in reference to the Antares system design. Two beam-alignment schemes which satisfy the technical and operational requirements and are cost-effective with respect to hardware and integration costs are outlined and compared. The first scheme, the Flip-In Detector technique, can use CO2 or visible alignment lasers. A series of beam centering detectors are sequentially centered at appropriate mirrors throughout the 72 beam lines. The beams are finally centered at the target position on a centering detector. The second scheme, the See-Through Imaging Technique, has a visible-imaging, electro-optic centering and pointing system whose optical axis is colinear with the front-end amplifier driver pulse. The imaging system can be centered on each mirror throughout the optical train. The target can be viewed and the beam pointed and focused at any part of the target.

The Antares laser system is a large (100 kJ) CO2 pulse laser operating at 10.6 pm. The system has 72 beam lines, each with an aperture of 900 cm2. The system will be composed primarily of large copper-faced mirrors whose principal dimensions range up to 65 cm. These mirrors will be single-point diamond turned (SPDT) at the Y-12 facility of Union Carbide Corporation in Oak Ridge, Tennessee. We have had to develop surface quality specifications for these mirrors. These specifications were initially set at 50 nm peak-to-valley (p-v) surface error for the microsurface over 0.5-mm areas and 500 nm (p-v) over the whole mirror surface. In this paper an attempt has been made to refine these specifications to a more phys-ically meaningful set based on the performance of the system. The optical specification for Antares is that 80% of the energy from each beam should be deliverable inside a 400-μm circle. The diffraction limited focal spot is 160 pm across, so small amounts of low spa-tial frequency wavefront aberrations are acceptable. This is the "figure error" and can be represented by a best-fit fourth-order polynomial. It is specified separately from the higher spatial frequency "subfigure" errors that diffract light out of the 400-μm circle. Antares will have a completely automatic alignment and centering system. A more versatile and less expensive alignment system can be developed if the alignment is done with visible light. This tightens the tolerances on the microsurface but not the figure error. These requirements, along with several lesser ones, must be considered when tolerancing the mirror quality. It appears that the SPDT mirrors turned at Y-12 will meet our minimum requirements.

We describe the optical systems of the six Antares Laser Power Amplifiers. These assemblies are preceded by the front-end optics and followed by the target system. Each power amplifier receives an annular input beam and divides it into 12 beams which are then directed to double pass them through 12 gain regions surrounding a central electron gun. Provisions are being made for spatially filtering each beam and for the possibility of adding saturable absorbers. Two keys to the successful completion of the power amplifier are: (1) the avoidance of unwanted lasing modes and hot spots in the wavefronts, and (2) the maintenance of alignment throughout the entire laser system, including the internal alignment of the Power Amplifier. The goal has been to minimize alignment problems by careful and simplistic design of the mountings, stressing modular assemblies and accessibility. We have succeeded in designing to average energy densities of 2.0 J/cm2 for salt windows and 3.0 J/cm2 on copper mirrors, while extracting the largest possible energy from the volume of gas which is electrically pumped.

Diffraction and aberration effects are calculated for the power-amplifier and target-system portions of the 100-kJ Antares laser fusion facility, using LOTS, a fast-Fourier-transform propagation code incorporating a model for saturating gain in CO2. Energy losses due to diffraction are found to be small compared to other losses. Diffraction "hot spots" usually typical of propagation at low Fresnel numbers are effectively suppressed in the Antares power amplifier by gain saturation. Taking account of diffraction and aberrations over the whole optical train, the code predicts a target focal spot that has 82% of its energy in a 150-μm-diameter circle, a result essentially identical to what would be expected of the final focus mirror alone.

The Laser Optical Train Simulation (LOTS) code has been developed at the Optical Sciences Center, University of Arizona under contract to Los Alamos Scientific Laboratory (LASL). LOTS is a diffraction based code designed to calculate the beam quality and energy of the laser fusion system in an end-to-end calculation.

It is essential to minimize the pointing and focusing errors at the focal plane for many applications involving infrared laser systems. In the case of the LASL CO2 laser fusion systems, with two beams in the Gemini System and eight beams in the Helios System, this is particularly important. The LASL Helios CO2 Laser Fusion System has eight 34-cm diameter beams emerging from the power amplifier and each beam is brought to focus by an off-aperture parabola (nearly 77.3-cm focal length) resulting in a nearly F/2.4 beam at the focal plane. The design tolerance at the focal plane for pointing accuracy is ± 25 microns and for focusing accuracy is ± 50 microns for this system. This paper describes an alignment scheme based on the use of the infrared Smartt interferometer' and compares the results obtained using this technique and the autocollimating Hartmann scheme2 in a laboratory setup duplicating the target chamber region of one of the beams of the Helios System. The results using the Smartt interferometer show that pointing accuracy of ± 12.5 microns and focusing accuracies of ± 15 microns are obtained at the focal plane of the system.

The Gemini two beam CO2 laser fusion system incorporates a complex optical system with nearly 100 surfaces per beam, associated with the generation, transport and focusing of CO2 laser beams for irradiating laser fusion targets. Even though the system is nominally diffraction limited, in practice the departure from the ideal situation drops the Strehl ratio to 0.24. This departure is caused mostly by the imperfections in the large (34 cm optical clear aperture diameter) state-of-the-art components like the sodium chloride windows and micromachined mirrors. While the smaller optical components also contribute to this degradation, the various possible misalignments and nonlinear effects are considered to contribute very little to it. Analysis indicates that removing the static or quasi-static errors can dramatically improve the Strehl ratio. A deformable mirror which can comfortably achieve the design goal Strehl ratio of ≥ 0.7 is described, along with the various system trade-offs in the design of the mirror and the control system.

Analysis at Los Alamos and elsewhere has resulted in the conclusion that deformable mirrors can substantially improve the optical performance of laser fusion systems, as the errors are mostly static or quasi-static with mainly low spatial frequencies across the aperture resulting in low order Seidel aberrations in the beam. A novel deformable mirror assembly (Fig. 1) has been fabricated with 19 actuators capable of surface deflection of ±20 microns. The mirror surface deflections are produced by a unique differential ball screw that acts as both a force and position actuator. The screw is driven by a stepper motor giving a surface positioning resolution of 0.025 micron. No holding voltage potential is required, and a piezoceramic element in series with each ball screw provides a ±1 micron amplitude high-frequency surface dither to aid the correction process. Mirror performance in terms of individual actuator influence function, cross-coupling, figure attainment, long-term surface stability as well as optical performance characteristics will be discussed.

We shall discuss three optical diagnostics of laser-produced plasmas. These systems are designed to study lateral energy flow and plasma motion. One system has already produced significant information on return-current heating; these results will be presented.

When v2+v4 of CF4 at 1066 cm-1 is pumped by the 9.4-µm CO2 laser, stimulated emission on the (v2+v4) + v2 transition produces many discrete laser lines in the region 605 to 655 cm-1. A comprehensive program of Doppler-limited absorption spectroscopy of CF4 has been carried out using tunable semiconductor diode lasers, and has led to a full understanding of the rovibrational energy levels involved in the laser process. The frequencies of 28 laser lines of 12CF4 have been measured with an accuracy of ±0.2 cm- 1, for 12C16O2 pump lines from P(14) to R(24). From the complete vibration-rotation analysis of the v2+v4 band, the pump and laser transitions have been identified. Using the spectroscopic constants determined in the band analyses, we can predict within ±0.2 cm-1 the laser lines to be expected from any given pumping frequency. All observed laser lines have been accounted for; in a few cases there is evidence for a relaxation of J-value and/or Coriolis sublevel in the upper state. Application of these results to improving the performance of the CF4 laser and for designing it to produce specific desired output frequencies is discussed.

Optical pumping of 12CF4 with a high brightness CO2 laser system has produced energy conversion efficiencies over 10%. The CF4 laser output will be discussed in terms of pump laser parameters such as frequency and energy and CF4 laser parameters such as pressure.

The possible development of a room-temperature, 16 μm CF4 laser optically pumped by 5 μm radiation from a frequency-doubled CO, laser has been iyvestigated. This laser would utilize pumping ofithe v3 + v4 band of the CF4 at 1916-1 cm with subsequent lasing to v3 levels at 1283 cm-1 where there is essentially no thermally excited population. A spectrophone instrument was used to measure absorption cross sections of CF4 at the second harmonic frequencies of 22 of the 10 µm CO2 laser transitions. Lasing experiments to detect transitions at 16 μm are described.

Measurements have been made of the absorption of CO2 samples enriched in 18O and 17O by tunable diode lasers in the wave-number region 662.2-666.1 cm-1. The fundamental Q-branches of the 16O12C18O, 7 16O12C17O and 17O12C17O molecules are well resolved, together with "hot" Q branches and various R- and P-branch lines of other isotopic species; notably 18O12C18O and 17O12C16O. The molecular constants for the Q branch 0110 ← 00°0 of 16O12C18O have been calculated by a least mean-square fit to lines J=1 to 16 to be: vo = 662.3717 ± 0.0003 cm-1, ΔB = 9.65 ± 0.02 x 10-4 cm 1; where vo has been fixed in absolute magnitude by assuming the - 16O12C16O line, 0220 ←· 0110 P7, as a reference standard at 662.325 cm . More accurate values for ΔB and ΔD can be obtained by using the separation between several R lines and adjacent Q lines, and the known positions of the R lines relative to the origin to obtain: ΔB = 9.685 ± 0.002 x 10-4 cm 1 and ΔD = 2.2 ± 0.1 x 109 cm -1. Other molecular constants can be determined with highest precision by using low J-, R-, and P-branch lines as calibration standards, together with etalon spacing. The calculated line positions of the fundamental Q branch can be used as secondary standards to fit the hot bands.

Brewster cut CdSe crystals pumped by the P1(8) line of an HF laser were used to amplify optical signals at 16 μm. The source of the signals was a CdSe OPO pumped by the P2(7) line from the same laser. The configuration of both pump beams was such that flat radial intensity profiles were produced at the respective crystals, eliminating the need for confocal focusing. The gain bandwidth was measured and found to be in reasonable agreement with theory. The second order nonlinear susceptibility was measured for the first time at 16 μm and found in fair agreement with values determined at shorter wavelengths scaled to 16 μm through Miller's Δ.

We report a completely grating tuned (1.9 to 2.4 µm) picosecond traveling wave it generator capable of controlled spectral bandwith operation down to the Fourier Transform limit. Subsequent down conversion in CdSe extends the tuning range to 10 to 20 µm.

The spectroscopic properties and several applications of a new nonlinear material developed for use with the CO2 laser are described. The material, which is a prototype of a potentially large class of such solids, is constructed by doping a normally transparent material, KC1, with an impurity ion, ReO4, to produce absorptions in the frequency range of interest to the CO2 laser. In this paper we discuss the spectroscopy of the Re04 ion in KC1 and then proceed to describe several applications of this material to the CO2 laser . Using this material, extracavity pulse compression from a CO2 oscillator has been demonstrated. Second, the material has been used intercavity to produce modelocking. Both of these results are described.

A flashlamp pumped liquid nitrogen cooled Ho:YLF laser was constructed and operated both normal mode and Q-switched. In normal mode operation, the laser produced over 300 mj at 2.9 percent slope efficiency but it operated on several lines. Lasing was restricted to one line and Q-switched operation was achieved up to 150 mj.

We report operation of a rare gas halide mini laser with an active area 10 cm x 2 mm x 4 mm and complete dimension including blumlein circuit slightly over one square foot. KrF lasing of greater than 320 milliwatts average power at 333 Hz is reported and steady lasing at over 1.25 kHz is obtained. The energy per pulse at low repetition rate is 1.8 mJ. XeCl lasing of greater than 60 milliwatts average power at 350 Hz is reported, and steady lasing at over 850 Hz is obtained. Pulsewidth control is obtained from 10 ns to over 50 ns via control of the cavity Q.

Experiments on output beam divergence for a 1-Joule KrF laser with M = 4 and M = 10 unstable resonators for 112.5 cm, 187.5 cm and 225 cm separation are presented. Results gave beam divergence no better than some 30 times diffraction limited in the best case.

We have demonstrated efficiencies approaching 100% for the translation (upward or downward) of a 10.6 μm laser beam by .31, .53 and 1.16 cm-1 using the linear electro-optic effect in CdTe. Equations are given for extrapolation of these results to other laser wavelengths and other amounts of shift.

We have performed analysis and experiments to demonstrate the feasibility of laser pulse shaping, based on the coherent synthesis of waveforms by phase locked oscillators. The approach is to control the frequency, amplitude and phase of a number of waveguide CO2 lasers and to synthesize pulses consistent with Fourier analysis. The laser frequency and phase are controlled with a crossover servo which drives a piezo-electric length tuner and an electro-optical crystal modulator. This approach has been demonstrated with a five-laser array producing pulses of a nominal 1.6 ns duration.

Recent spectroscopic data is used to calculate small signal gain and absorbtion coefficients for the P and R branches of the 9 micron and 10 micron transitions in CO2. It is shown that in order to accurately calculate such quantities one must include the effects of overlapping lines and near coincidences due to higher lying vibrational transitions. Comparison with measured gain and absorbtion coefficients shows agreement to the level of several percent.

The generation of ultrashort CO2 pulses by optical free-induction decay in KCl:KReO4 is described. Here the narrow v3 vibrational mode is coincident with the 10.6 μm, P(26) laser line. For instantaneous shuttering of the incident laser beam, pulses in the picosecond range may be produced by this material. In practice, the minimum pulse length and pulse intensity were severely limited by the breakdown time of the nitrogen plasma cell.

The design and performance of a uv preionized CO2 oscillator capable of 10 atm operation is described. When used as a reinjection oscillator this device is capable of subnanosecond pulses of CO2 radiation with powers in excess of 1 GW.

With multipulse energy extraction in electric discharge excited CO2-laser amplifiers, we can convert a significantly higher fraction of the vibrational energy stored in the gas by the discharge into laser light. The time scale for efficient energy extractions is determined by the collisional transfer rates for the flow of vibrational energy into and among the various vibrational modes of the gas molecules. We briefly survey the collisional processes of interest, describe optical techniques of implementing the multi-pulse method, and give calculated results for a representative case. The results indicate that high laser efficiencies, attractive for future commercial applications, can be achieved with CO2 lasers.

A new class of position-sensitive detectors has been developed at Los Alamos Scientific Laboratory (LASL). These detectors utilize Seebeck effect and thermal diffusion in the detector material (Si). Large area (up to 125 cm2) detectors with only four electrodes yield accurate x-y position sensing for cw and pulsed laser beams. The detectors are sensitive to any laser wavelength and feature high optical damage thresholds (>109 W/cm2 at 10 μm). The detector responses are found to agree accurately with a thermal diffusion model that has no adjustable parameters. These detectors are currently in use on Helios and are planned for use on Antares, the LASL laser fusion systems.

Sub-one-hundred-ps pyroelectric detectors are required for laser fusion and fast-time spectroscopy experiments. A Los Alamos Scientific Laboratory program is evaluating commericially available detectors and detector materials to develop a pyroelectric detector capable of fast operation without damage and compatible with the 5-GHz oscilloscope direct access mode of operation. The latter requires approximately 1-V signal levels across 50 Ω. The first phase of this program has been spent in developing a fast detector design. The fall time of the detector is approximately 30 ps, limited by a stray capacitance of 0.14 pf. The rise-time is less than 15 ps. Further improvements in the response time require reducing the stray capacitance. The second phase of this program is to measure the damage threshold of various pyroelectric materials and devices, evaluating several surface treatments and device preparation methods. Some of the materials under consideration include strontium barium niobate, lithium tantalate, lithium niobate and lanthanum-doped lead zirconate. Preliminary results will be presented.

Two of the basic types of diffused junction silicon photodiodes are the n/p diode (which can be operated in either a partially-depleted or fully-depleted mode) and the p/n diode (which is usually operated in the unbiased mode). These two types of diodes are illustrated in Figure 1. While both diodes convert incident photons to electrical current by the same process they exhibit entirely different operating and performance characteristics which determine their suitability for particular applications. Several of these characteristics are reviewed below.

The Multi-Anode Microchannel Arrays (MAMAis) are a family of photoelectric, photon-counting array detectors being developed for use in instruments on both ground-baased and space-borne telescopes. These detectors combine high sensitivity and photometric stability with a high-resolution imaging capability. MAMA detectors can se operated in a windowless configuration at extreme-ultraviolet and soft x-ray wavelengths or in a sealed configuration at ultraviolet and visible wavelengths. Prototype MAMA detectors with up to 512 x 512 pixels are now being tested in the laboratory and telescope operation of a simple (10 x 10)-pixel visible-light detector has been initiated. The construction and modes-of-operation of the MAMA detectors are briefly described and performance data are presented.

This paper describes the performance characteristics of the following four new types of subnanosecond optical detectors: (1) a triplanar phototube with 100-150 ps FWHM response and less-than-unity gain, (2) a microchannel plate photomultiplier tube with 300-500 ps FWHM response and up to 106 gain, (3) a magnetically-focused electrostatically-deflected streak tube with a 10 ps time resolution capability and (4) an all-electrostatic circular-scan streak tube with a 7 ps time resolution capability. The first two detectors convert a fast input optical pulse, or chain of pulses, directly to a fast output pulse (or output pulse chain) in a coaxial transmission line, while the latter two detectors convert a fast input optical pulse to an intensity-modulated spatially-distributed output optical image.

Several standard plasma and gas dynamic diagnostic techniques have been integrated into a system for observing the formation and propagation of high-power Nd:glass-laser generated one- and two-dimensional shockwaves in air from 0.1 torr to atmospheric pres-sures. Diagnostics include either single-frame, two-wavelength holographic ruby-laser interferometry or single-frame, single-wavelength interferometry with ten frames of shadow-graphy. Streaks or ten frames of the early luminous shocked region also are taken on all shots, as well as time-resolved luminosity measurements using high-speed biplanar vacuum photodiodes with various wavelength interference filters. Shadowgraphy frames are 200-ns long at 1-μs intervals, while emission frames are variable with a maximum 10-ns exposure and 50-ns interval. Both the streak mode and emission measurements with the vacuum diode allow subnanosecond time resolution. The interferometry provides 20-ns exposures from 500 ns to late times. Methods for reducing and interpreting the data have been, or are currently being, developed. Interactive computer programs for digitizing the fringe patterns provide fringe-shift profiles for Abel inversion. This has provided neutral gas and electron density information in the spherical, one-dimensional cases. Diagrams and photographs of the experiment will be shown as well as examples of the data that have been taken. Methods for data reduction will be outlined and some of the results shown.

The advantages and applications of optical fiber data transmission in radiation diagnostics are considered. The individual components of such a system are each discussed. Only passive systems utilizing a direct radiation-to-light conversion are studied. The development of new scintillators with long wavelength emission characteristics was critical to such application and scintillators with a FWHM of 1.4 ns and significant output at 600 nm have been achieved. Cerenkov converters also are used in high bandwidth systems. System examples using scintillators are given, which show a 12 exceeding 100 MHz over f distances of almost 200 m (with a dynamic range of 4 x 104) or 500 m (with a dynamic range exceeding 100). With a Cerenkov converter, system examples show a bandwidth approaching 1 GHz over 1 km fiber lengths. Experimental data that illustrates both systems will be shown.

Fiber optic technology is being used to solve electromagnetic compatibility problems on ZT-40. Design and performance parameters of four different types of optical links are discussed.

The application of low loss multimode optical fibers to nuclear diagnostics has been discussed in previous papers. Fiber requirements for this application differ substantially from those for normal communications use. The emphasis for nuclear measurements has been on development of high frequency analog fiber optic transmission line systems, which range from 100 MHz to >500 MHz signals transmitted at 600 nm and 800 nm, respectively. Accordingly, specialized fiber characterization procedures over a wide spectral range have been developed. These techniques include measurement of material and modal dispersion, optical attenuation, and optical linearity. It is also important to know the prompt radiation response of optical fibers in nuclear diagnostics. Late time measurements of this type have been discussed in a previous paper. We will discuss a system to improve these measurements to the subnanosecond regime.

The cumulative baseband frequency response of interconnected fiber optic cables is modeled as tandem networks. Three models are described based on the degree of intermodal coupling. Measurement results are presented for verification of the models.

We have developed a general treatment of light propagation and dispersion in arbitrary refractive index profile optical fibers. The method is based on the accurate numerical solution of the scalar wave equation by a combination of operator splitting and Fourier transform techniques. The Fourier representation of the propagating beam allows a dual description of the electric field in both (x,y) and transverse wavenumber space. This dual description affords a precise characterization of such propagation properties as spatial beam confinement, angular confinement (numerical aperture) and losses, all as a function of propagation distance z. In addition, Fourier transformation of the field autocorrelation function with respect to z yields the complete power spectrum containing trapped, "leaky," and radiation modes. The trapped modes appear as distinct spectral lines whose position (modal wavenumber) and amplitude (modal weight) are the ingredients needed to characterize the dispersion of the light pulse. The numerical tools developed allow a general characterization of realistic refractive index profiles including, e.g., losses and fiber bends as appropriate. The method has been applied to a fiber having a square law refractive index profile with a central dip. The cladding was surrounded with a strong absorber so that leaky modes would be rapidly attenuated. The central dip is found to profoundly alter the mode spectrum, principally by removing some of the degeneracy of the square law modes. In addition, the spread of modal group delays is increased by nearly two orders of magnitude, leading to much less favorable dispersive properties.

We have generated phase-conjugate, 10 μm reflection from the grating established with counterpropagating waves in both germanium samples located within an oscillating TEA CO2 laser cavity and in the CO2 gain medium itself. The work in germanium is the first demonstration of nonlinear phase conjugation in the infrared and of intracavity degenerate 4-wave mixing. The work within the CO2 gain medium is the first demonstration of 10.6 μm phase conjugation in an inverted medium. In our first experiments we used a hybrid, single line, TEA laser with a peak output intensity of 2 MW/cm2. A wedged germanium flat output mirror was reversed so that the Ge coupler, internal to the cavity, served as the nonlinear medium. This intracavity technique eliminated the difficulty of precisely aligning a pair of counter-propagating beams in the nonlinear sample. Using an infrared vidicon, we demonstrated that the placement of an aberrator in the beam line had little effect on the germanium reflected signal, whereas the same aberrator in front of a reference mirror caused substantial distortion.

During the study of focusing and phase conjugation of photon echoes it became apparent that similar effects occur for either simultaneous pulses or continuous laser radiation. These effects are apparent in the simple two level atom theory in which the stimulated electric dipole is proportional for steady state to [EQUATION] where the electric field amplitude E = A eiφ + B . A(x,z) and φ(x,z) are the amplitude and phase of the signal wave and B is the amplitude of a plane carrier wave. The phase conjugate wave A e-1φ is generated by the A2 + B2 + 2 AB cosφ term in the denominator. Phase conjugation is limited to physically thin cells and a 1 mm Na cell limits the angular spread to ± 20 milliradians. Current experiments with a 2 mm thick Na cell will be discussed. For two laser beams at angle θ, radiation at the phase conjugate -θ, and at higher angles ± 2θ, ±3θ, etc. are observed. Limited phase conjugation such as focusing of a curved wavefront has been observed. More extensive phase conjugation results will be reported.

The Abrams-Lind formulation of degenerate four-wave phase conjugation is generalized to treat signal-pump detuning. The theory is an extension of two-wave grating-dip spectroscopy. We find that the reflection passband of phase conjugation is limited by the power-broadened bandwidth of the population difference, a width that can be very narrow. Unlike the running-wave saturator case, the signal absorption saturated by a standing wave shows no gain. This is due to the presence of appreciable amounts of unsaturated medium between spatial holes. We also generalize the degenerate three-wave mixing coefficients of Heer to include signal-pump detuning.

The focal spot size and intensity of CO2 laser beams which are used for fusion studies is degraded because of manufacturing defects in the optics train and because of alignment errors. Phase conjugation can in principle provide diffraction-limited performance despite imperfect optical components. This paper describes how such a system might be adapted to existing laser systems and discusses the problem areas which may be encountered.

We describe an optical system which produces two simultaneous, time-gated, magnified images of the back-side of a planar laser target. The two views of the target are from different directions relative to the target normal but may be chosen to be at the same angle relative to the direction of incidence of the laser beam so as to permit cancellation of any angular correlation relative to the laser beam direction. The two images are amplified by a time-gated channel-plate intensifier (CPI) and recorded on 35 mm film. This device permits measurement of the angular distribution relative to the target normal of light emitted from the back-side of laser targets as a function of time relative to the start of the laser pulse; such measurements are required in an experiment described in the report, "The Limb-Darkening Opacity Experiment Using a Laser-Heated Plasma," LA-7484-MS. We describe the operation of the device and discuss an important instrumental effect due to non-simultaneous gating of different portions of the CPI tube whereby an incorrrect angular distribution can be obtained. We discuss precautions necessary to insure that this does not occur. Samples of data will be presented.

A loss of resolution was experimentally observed when using a 40-mm microchannel plate image-intensifier tube at high gains. A controlled set of experiments was performed to study this effect. A resolution chart, under constant illumination, was viewed while a lens-image intensifier combination was varied so as to maintain constant light output from the system. Significant resolution degradation can occur at the higher gains. Experiments were performed to determine whether this phenomenon was due to photon statistics at the photocathode or to an increase in noise at the higher microchannel plate (MCP) voltages. The experiments showed that the controlling factor was the voltage across the MCP and not the photon statistics. A theoretical calculation is presented that confirms and augments the experimental results.

A fast solid-state camera has been designed to record diagnostic pictures of unpredictable high-speed events. The sensor is an array of silicon photodiodes that are read out in parallel. The system design features a variable framing rate to 100,000 frames/ second, continuous recording with asynchronous stop trigger, a solid-state memory holding 256 frames, and electronics having a dynamic range exceeding 2000:1. The image data is retained in a circular memory permitting the camera to run continuously until the event occurs. Immediately afterward the camera is shut down with a memory of 256 frames covering the event. The camera can be synchronized with similar cameras covering different aspects of the same scene. Operation of the camera is monitored and controlled by a microcomputer. The operator can select framing rate, image presentation on the CRT, and synchronization of multiple cameras.

The novel spectrometer here scans the path difference that is introduced between two beams that are produced with wavefront division by variation of angle of incidence on a reflection lamellar grating of fixed groove depth--not variation of groove depth at normal incidence, which is impractical for the near infrared owing to friction. Performance of a shallow groove grating is presented, and performance of a deep groove grating is predicted. An application of transparent lamellar gratings, with telescope objectives, is described.

An all-mirror optical system is used to direct the light from a variety of spectroscopic sources to two 2-m spectrographs that are placed on either side of a sturdy vertical mounting plate. The gratings were chosen so that the first spectrograph covers the ultraviolet spectral region, and the second spectrograph covers the ultraviolet, visible, and near-infrared regions. With the over 2.5 m of focal curves, each ultraviolet line is available at more than one place. Thus, problems with close lines can be overcome. The signals from a possible maximum of 256 photoelectric detectors go to a small computer for reading and calculation of the element abundances. To our knowledge, no other direct-reading spectrograph has more than about 100 fixed detectors. With an inductively-coupled-plasma source, our calibration curves, and detection limits, are similar to those of other workers using a direct-reading spectrograph.

The remote high resolution infrared imagery of the shuttle orbiter lower surfaces during entry to obtain accurate measurements of aerodynamic heat transfer will occur on the first shuttle flight. These images will have a spatial resolution of 1 meter or better and a temperature resolution of 2.5% for temperatures between 800 and 1900°K. The experiment is accomplished utilizing an aircraft on station along the predicted shuttle entry ground track at an altitude of 14Km and having on board a telescope with an aperture of about 100 cm. Tracking of the shuttle, prior to encounter with the main on board telescope, is necessary to correct for shuttle entry uncertainties, aircraft position uncertainty and platform drift. Once acquisition of the shuttle is accomplished, a closed loop servo takes the system to complete encounter and the subsequent infrared images as the shuttle passes through the field of view of the 91.5 cm Kuiper Airborne Observatory operated out of NASA Ames Research Center.

The morphological parameter of cell size has been extracted from static light-scatter patterns of single biological cells illuminated with a He-Ne laser. This was done using fabricated biological cell models. Once calibrated, size measurements can be performed on either biological cells or objects by Fourier transforming their light-scatter data.

The need for an x-ray imaging system with better than micron resolution is increasingly evident to workers in laser fusion and general plasma diagnostics. As early as 1974, John L. Emmett of the Lawrence Livermore Laboratories in an article titled, "Shopping List for Fusion" cited the need for hardware to make measurements with micron resolution in the soft x-ray region. In the same year, KMS reported that their x-ray image of an imploding pellet, formed with a pinhole, had a resolution of only 14 microns. Laser-fusion diagnostic requirements are such that it requires viewing a small object at neither very small or very large distances. What is needed is an x-ray microscope with a large working-distance objective. For this and other reasons we favor the basic Kirkpatrick-Baez arrangement. We shall explore some of the problems and solutions associated with the formation of x-ray images at small angles of grazing incidence mainly in the crossed-mirror configuration.

A typical sample that can be accommodated in a high-pressure diamond-anvil cell is about 300 μm in diam and 100 μm thick. In this microscopic working volume we observed phase transitions of various condensed gases. Pressures were determined from measurements of fluorescence line shift in a ruby chip imbedded in the sample. We observed freezing in the gases and we conclude that solid phase transitions can be detected by the sudden changes they cause in depolarization of transmitted light, even when viewed through stressed diamond anvils.

A simple interferometer was built to fit inside a 0.64-cm bore high-pressure cylinder. One end of the cylinder was sealed with a sapphire optical window having a clear aperture of the unsupported area of a Bridgman seal. As the gas density in the cell is increased, successive interferences from a He-Ne laser are detected with a simple photo diode. From a calculated value of the refractive index at the He-Ne wavelength and a measurement of pressure against a calibrated pressure gage at a known temperature, the order shift with pressure is determined. On the basis of an assumption that the Clausius-Mossotti relation is constant, we calculate dP/dP and make comparisons with the equation of state. Measurements and comparisons for nitrogen up to 200 bar are presented and shown to agree well with the derivative values, dp/dP, obtained from the National Bureau of Standards equation of state. One order of interference represents a density change of p = 1.36x 10-4 gm/cm3 for N2 gas at 1 bar. In addition to the usefulness in equation of state studies with gases, the method permits calibration of phase-sensitive optical instruments and may be useful for the assessment of laser target gas loading.

The retinex is the name given to each of four independent systems. Each of these uses a liaison between the whole area of the retina, the pathway to the cortex, and the cortex to generate what we call objects in the outer world. Each of the systems is rigorously independent of the others although the band of wavelengths used by each overlaps rather broadly the bands used by the others. The colors of the objects generated are determined by the comparison, presumably cortical, of the four different constructs produced by the four retinexes. Each of the retinexes associated with the left eye has a sister retinex associated with the right eye, and these pairs impose a geometric rigidity on the space created by the retinal-cortical system.

An incoherent optical system that performs the product of an input vector and a stored matrix is described. The input vector is suitably biased and entered into the system on an array of infrared light emitting diodes. The output vector is measured on parallel elements of a PIN photodiode array. With proper data encoding, vectors and matrices with complex elements can be achieved. An experimental system designed to perform discrete Fourier transforms is described. While the throughput rate of the current system is 108 data samples per second, systems with throughput rates of 1010 can be envisioned.

Recent dramatic increases in the cost/performance ratio of computers are beginning to have some impact on the field of optical design. Although the mainstream of optical design still utilizes large computers, production programs that run on minicomputers are now well established, and we have written programs of somewhat limited scope that run on desktop systems.

The optical design of short-pulse CO2 laser-amplifier systems depends upon five considerations. First is the design for laser output, viz., the desired energy, power, frequency spectrum and pulse shape. These specifications determine the pressure, volume, and gain coefficient of the final amplifier stages, and the characteristics of the oscillator-switchout section. Second, constraints imposed by damage limits of the optical components, especially the output windows and the large beam relay and focus mirrors, fix the minimum laser aperture. Since the performance is limited by the maximum, rather than the average, optical power or energy, the gain should be as uniform as possible across the aperture. Third, parasitics must be suppressed. Although the limits on the small-signal gain which can be supported in CO2 amplifiers have increased with the discovery of nonlinear absorbers and optical blacks at 10-μm wavelength, parasitics always occur if the gain exceeds some limiting value. The value of the allowed small-signal gain influences the pressure and volume of the amplifiers through the interrelation with stored energy and pulse amplification. Fourth, the optical layout of the entire system must be designed to accommodate the considerations mentioned previously, as well as additional systems requirements for multipass energy extraction, efficiency, contrast ratio, beam expansion, aberrations, attenuation of target backscatter, and provision for alignment and focus. Finally, no matter how well the laser has been designed, the performance will be substandard if the optical components are not, or cannot, be manufactured to specifications. The LASL Helios and Antares lasers will be described to illustrate the influence of these considerations on the optical design of the present and future high-performance CO2 lasers, which are necessary for the development of laser fusion. Work performed under the auspices of the US Department of Energy.

Since the discovery and the first demonstration of high power laser action in CO2, the gas lasers have become free of the criticism of their being only low-power systems. The CO2 laser, which has been followed by numerous other high-power gas laser systems (perhaps not with the same power capability and ease of operation as the CO2 laser), has opened up the infrared region for a large number of scientific and practical studies. In this paper, I will describe in detail one specific application--that of optoacoustic spectroscopy in both gaseous as well as liquid state. In addition, I will attempt to provide pointers for future directions for improvement of high-power gas lasers and possible new applications of the old as well as not yet invented gas lasers.

One of the many areas in which lasers are beginning to play an important role is that of chemistry. One application of great significance is the use of low average power tunable lasers as an analytical tool to measure or study reaction kinetics and products. The other application of equal or greater importance is the use of high average power tunable lasers as a manipulative tool to control or drive chemical reactions in industrial processes. Here the cost of photons is the most important factor and a new device called the "Free Electron Laser" is under development which promises to reduce the photon cost to less than one dollar per pound of material, thus including a wide range of industrial products.

Nd:glass lasers have been developed to a high degree of performance and reliability. At the Lawrence Livermore Laboratory this development started very early with the long path laser, progressed to the new operating Shiva laser, and reached the present state of the art with the Nova laser design. Laser power has increased from 0.04 TW on early lasers to > 20 TW on Shiva, to a projected > 300 TW performance on Nova. Energy has similarily increased from ~0.07 kJ on early lasers, to > 10 kJ on Shiva, and > 300 kJ projected on Nova. The techniques developed to support this technical progress are summarized in this talk. A great deal of this work is not specific to Nd:glass lasers and is useful to all workers concerned with fusion lasers. In the spirit of this conference, emphasis will be placed on issues of propagation, optical quality, damage, and laser architecture.

In the present paper, we will first present a boundary-type solution for total internal reflection at an interface between two nonabsorbing media, in which the instantaneous, time varying and time averaged radiant fluxes have been determined at all points in the two media. Solutions for the s and p polarizations were found for which the instantaneous tangential E and H components and the normal components of the radiant flux were continuous in crossing the interface. From these radiant fluxes, it was possible to derive equations for the flow lines, to determine the instantaneous radiant fluxes along these flow lines, and to see how the methods of propagation differed in the two media and for the two polarizations. These ideas have also been extended to long wavelength reflection at ocean-air interfaces, and similar phenomena appear there.

This paper discusses three recent interferometric optical testing studies being performed at the Optical Sciences Center. First, the design and construction of a 10 μm infrared Twyman-Green interferometer is discussed. The basic specifications of the interferometer are described and results are shown for testing optically rough surfaces having an rms surface height variation as large as 1.38 μm. Second, the use of an infrared point-diffraction interferometer is described and results are presented for testing a germanium lens. The effect of a finite pinhole size on the phase and amplitude uniformity of the reference wavefront is discussed and results are shown for an astigmatic wavefront and a spherically aberrated wavefront. Last, a video interferogram processor being built for analyzing both visible and infrared interferograms is described and interferogram analysis capabilities are discussed and illustrated.

Measurements of the atmospheric modulation transfer function (MTF) along a vertical path are presented for both mountain and desert locations. The mountain and desert sites are found to be equivalent during the day but, at night, the mountain sites have twice the resolution observed at desert locations. The atmospheric MTF above the desert at night is found to be essentially the same as during the day. Measurements of the optical turbulence in the first kilometer above the surface are shown to account for these findings.

The emphasis in this minicourse will be on the theory and experimental verification of topics in short-pulse propagation in high-pressure, short-pulse CO2 laser amplifiers. We begin by deriving the basic equations of a two-level laser amplifier followed by examining the conditions under which a coherent electric field or incoherent rate equation description applies. Some exact solutions of the two-level system will be discussed. Finally, we discuss the results of this type of modeling applied to our LASL high power, short-pulse CO2 amplifiers.