Electronically Scanned Waveguide Laser Arrays V. J. Corcoran and I. A. Crabbe

Both authors were with the Martin Marietta Corpora­tion, Orlando, Florida 32805, when this work was done; V. J. Corcoran is now with the Institute for Defense Analyses, Arlington, Virginia 22202. Received 6 March 1974. Since the waveguide laser concept was first suggested,1

waveguide lasers have operated in the visible,2 ir,3 and far-ir regions.4 Laser-pumped far-ir laser action has also been reported.5 The waveguide laser makes available a high energy density source of coherent radiation. An ex­tension of the waveguide laser concept to a phased array of waveguide lasers that can be scanned electronically is presented here. An electronically scanned laser array can produce power densities comparable with those of a chemi­

cal or gas dynamic laser. In addition, the power density can be moved in space at rates associated with electronic rather than mechanical scanning.

The electronically scanned laser array employs rows and columns of waveguide lasers with an electrooptic phase modulator placed in front of each laser to adjust the phase of one laser relative to another. For example, single crys­tal CdTe modulators can be used with CO2 lasers. Opti­cal feedback from the lasers is used to injection lock one laser to another so the relative phase of each laser is con­stant. The inputs to the modulators then control the rel­ative phase of the lasers. This, in itself, can be used to move the beam in space by adjusting the phase of each laser modulator electronically. In this way a rapid scan can be achieved, since the response time of an electro-optic modulator is very fast compared with mechanical' scanning.

Although the array as described above can be electroni­cally scanned, its accuracy is impeded by the possibility of drift in frequency of the laser. This can be overcome by the extension of the phase lock technique reported for the HCN laser6 to the wavelength of interest. ': For exam­ple, with this method a signal is multiplied from a stan­dard frequency to the frequency of the CO2 laser. This signal is used as a reference for a phase detector, so the error between the CO2 laser phase and the reference is used as an error signal to correct the phase of the laser through the laser power supply. As a result, an absolute­ly stable laser signal that can be adjusted in phase can be used for each element in a scanned array because each el­ement is properly phased to each other through injection locking and phase modulation (see Fig. 1).

A waveguide laser with a 1-mm bore diameter and a 7.6-cm length can produce an average output power of more · than 2.5 W with a 6-10% efficiency. The cross sectional area of the device can be made small, even when a cooling jacket is used. Consequently, it is an easy task to obtain an array of 104-105 elements in a small package. The output power of such an array would be competitive with a gas dynamic or a chemical laser.

The CO2 laser array described is a highly thinned array with an aperture approximately 100 times the wavelength

August 1974 / Vol. 13, No. 8 / APPLIED OPTICS 1755

Fig. 1. Transmitter detail showing injection locking lasers, phase locking of the laser array to a frequency standard through the power supply, and scanning effected by driving phase modulators

through a computer.

of the radiation and a spacing of approximately 500 times the wavelength. If sources are equally spaced elements in the array, then closely spaced grating lobes exist. These grating lobes would result in the waste of energy and an ambiguity in a target location when the array is used as a radar transmitter.

The grating lobe problem can be circumvented by mak­ing the spacing between elements unequal. For example, consider an array with an element at the origin and ele­ments symmetrically located about the origin. The far field intensity pattern, which is related to the Fourier transform of the aperture pattern is given by

where λ is the wavelength, b is the aperture width, θ is the angle from the normal to the array, and dn is the dis­tance from the center of array to the center of the nth ap­erture. This equation is a product of an aperture function and a grating function. A method for reducing the grat­ing lobes comes from inspection of the grating function: to chose the spacings to be nondegenerate, i.e., the spac-ings should be such that the grating lobes due to one spacing do not coincide with another set of grating lobes. This occurs when the ratio of any two spacings is not an integer. In this case, the main lobe intensity is propor­tional to N2, and the grating lobes are down 40 dB from the main lobe for N = 200.

A computer program has been used to check the feasi­bility of reducing the grating lobes as indicated by the above equation. The computed far field pattern for a symmetric linear array with 91 elements has been plotted in Fig. 2. Each element has a linear aperture of 2 mm, and the pseudorandom spacing is approximately 1 cm be­tween elements. The spacings chosen and the number of elements used result in significant reduction of the grating lobes. These data are the first attempts at reduction of grating lobes with pseudorandom spacings. Significant improvement could be expected with further judicious choices of separation.

For any array, scanning can be achieved by adjusting the phase of each waveform from an aperture so the wave­forms add at the desired point in space, i.e., so the optical path length to that point from each aperture is an integral multiple of a wavelength. For the symmetrical array, the

Fig. 2. Computer run for far-field pattern of 91-element symmetrical array with unequal spac­ing between elements; λ = 10 μm; b = 1 mm;

and dn ~ 0.5-1.0 cm.

1756 APPLIED OPTICS / Vol. 13, No. 8 / August 1974

objective is to choose path distances such that any pair of elements produce grating lobes spaced differently from any other pair as was done for θ = 0. This can be achieved by choosing the phase adjustment of one element of the nth. pair as φn1

= —kdn sinθ and the other element as (φn2

= + kdn sinθ. The intensity is then a maximum at θ. To avoid the requirement to maintain the grating lobes at a low level continuously, it may be necessary to use a continguous digital electronic scan rather than a continuous scan.

The electronically scanned laser array, when coupled with an optical receiver, could be used as a track-while-scan laser radar. When used with heterodyne detection, long-range detection is possible because of the potentially. high power densities. Since the power densities are com­parable with that attainable with chemical lasers or gas dy­namic lasers, the device might also be useful as a weapon. The combination of a track-while-scan radar and a weap­on from one package is conceivable.

The laser array offers the advantages of low power den­sity at the source because the outputs are essentially par­allel. Also, a high degree of coherence in a single mode is possible, which may not necessarily be possible with gas dynamic lasers or chemical lasers. The waveguide lasers can be sealed so no fuel or exhausts are necessary. The rapid scan rates possible are not attainable with large mirrors that are used at high power levels in gas dynamic lasers and chemical lasers. Finally, this system elimi­nates the need for large optics and optics that must be ca­pable of withstanding high power levels while retaining their optical characteristics.

References 1. E. A. J. Marcatili and R. A. Schmeltzer, Bell Syst. Tech. J.

43, 1783 (1964). 2. P. W. Smith, Appl. Phys. Lett. 19, 132 (1971). 3. T. J. Bridges, E. G. Burkhardt, and P. W. Smith, Appl. Phys.

Lett. 20, 403 (1972); and E. G. Burkhardt, T. J. Bridges, and P. W. Smith, Opt. Commun. 6, 193 (1972).

4. H. Steffen and F. K. Kneubuhl, Phys. Lett. 27A, 612 (1968); P. Schwaller, H. Steffen, J. F. Moser, and F. K. Kneubuhl, Appl. Opt. 6, 827 (1967).

5. D. T. Hodges and T. S. Hartwick, Appl. Phys. Lett. 23, 252 (1973).

6. R. E. Cupp, V. J. Corcoran, and J. J. Gallagher, IEEE J. Quantum Electron. QE-6, 241 (1970).

August 1974 / Vol . 13, No. 8 / APPLIED OPTICS 1757