JPH0573075B2 - - Google Patents

Description

Claim 1: A method comprising the steps of: forming a plurality of corrugated portions on the surface of a substrate projecting outwardly from the remaining surface thereof; and a component to be deposited and a component serving as a solvent for the substrate material; creating a first solution exhibiting a supersaturated growth condition on flat and concave surfaces of the substrate and a slightly unsaturated growth condition on the convex surface of the substrate; and a surface of the substrate having corrugated portions. is brought into contact with the first solution to partially dissolve the convex portions of the corrugated portion to form a land portion between the concave portions of the corrugated portion, and a first solution is applied onto the concave portion and the land portion of the corrugated portion. a step of causing the deposition of a layer until the first layer forms a flat surface; contacting the substrate on which the first layer is formed with another solution; cooling a substrate and depositing an active layer on the flat surface of the first layer;
Array manufacturing method.

2. The flat solution according to claim 1, wherein the step of preparing the solution includes the steps of equilibrating the first solution at an equilibrium temperature and cooling the first solution to a lower temperature. A method of manufacturing a laser array having an active layer.

3. The method of manufacturing a laser array with a planar active layer according to claim 2, characterized in that the substrate is at a lower temperature than said above.

4 The substrate is GaAs and the solution is Al, GA and
4. The method of manufacturing a laser array having a flat active layer according to claim 3, characterized in that it contains As.

5. Fabrication of a laser array with a flat active layer according to claim 4, characterized in that the lower temperature at which the solution in contact with the flat part of the surface becomes saturated is between about 700°C and 800°C. Method.

6. The method of claim 5, wherein said lower temperature is about 760°C.

7 The difference between the equilibrium temperature and the lower temperature is about 2°C to
10.degree. C. and the cooling rate is approximately 0.5.degree. C. to 5.degree. C. per minute.

8. A laser array with a flat active layer according to claim 7, characterized in that the difference between the equilibrium temperature and the lower temperature is about 4°C to 5°C and the cooling rate is about 1°C per minute. manufacturing method.

BACKGROUND OF THE INVENTION Semiconductor lasers generally consist of a body of semiconductor material with a thin active region formed between cladding regions of opposite conductivity type. In order to increase the output power from such a laser, a guide layer with a reflection coefficient intermediate between that of the active region and that of the cladding layer is inserted between one of the cladding layers and the active region. ing. The light generated in the active layer propagates through both the active layer and the guiding layer to form a larger beam at the emitting surface of the body. The thin active layer limits transverse vibrations, ie vibrations perpendicular to the plane of the layer, to the fundamental optical mode. In the lateral direction, in the plane of the layers, and in the direction perpendicular to the line between the laser planes, similar constraints do not exist and vibrations generally occur in several different optical modes simultaneously.

It has been found to be effective to introduce lateral variations in the laser structure to create an optical waveguide that confines the vibrations to the lateral fundamental optical mode. A channeled substrate laser formed by liquid phase epitaxy on a single channel in the substrate is
It is formed by a lateral variation in layer thickness and has an optical waveguide in close proximity to the absorbing substrate on the side of the emitting region on the channel. However, the lateral flow of current is not automatically restricted to the radiation area on the channel, but rather tends to flow towards the substrate at the sides of the channel. To limit this lateral current flow, reverse biased PN junctions are typically formed in the base or top layers on the sides of the channel. In U.S. Pat. No. 4,347,486, incorporated herein by reference, Botez discloses a laser having a pair of channels in the surface of a substrate with a mesa between them.
The layer overlying this channeled surface varies in thickness laterally because it tends to grow rapidly by liquid phase epitaxy on concave surfaces as opposed to planar or convex surfaces. This configuration limits the flow of current to the area above the mesa, and because the layer thickness varies laterally, the oscillations are essentially lateral across the mesa, to output powers of over 40 milliwatts. direction) mode.

Spaced apart separations in which the oscillation modes of the individual lasers are combined together to form a single phase-locked combined oscillator to increase the power of the coherent light beam beyond the capabilities of the individual lasers. Monolithic arrays of shaped laser devices have been proposed. Such lasers are pulsed and
an array of mesa-shaped waveguide lasers, the emitting region of which covers a mesa on the substrate; . The array operates in a phase-locked mode during pulsed operation, but is only partially phase-locked during continuous wave operation. In U.S. Pat. No. 4,385,389, Botetsu describes a phase-shift laser system consisting of a plurality of discrete laser elements of the type shown in U.S. Pat. Shows locked array. In this array, the lateral radiation leakage of the individual devices is large, so that coupling between the oscillation modes of the individual elements of the array occurs over relatively long distances. However, the use of channel pairs and the large spacing between the internal elements of the Botetsu array required due to layer curvature result in a far-field pattern.
This is not preferable because the number of lobes increases. Therefore, the spacing between the radiating elements is minimal and
It would be desirable to provide a phase-locked laser array that operates with a single narrow beam that peaks at 0 degrees.

SUMMARY OF THE INVENTION A phase-locked laser array includes a substrate having a plurality of channels substantially parallel to its surface. First clad layer, guide layer
A cavity layer consisting of an active layer and a second cladding layer are sequentially overlaid on the surface of the substrate and on the channel. A wide area electrical contact is provided on the second cladding layer above the channel.

The individual oscillators on each separate channel are coupled together primarily by the superposition of their instantaneous electric fields within the guide layer. Waveguides formed on separate channels suppress oscillations of higher order lateral modes over a wide range of outputs. By providing a large area of electrical contact, it is possible to provide a sufficiently uniform current distribution across the channel without having to confine the current to an area on the channel and without appreciably increasing the threshold current. .

The invention also includes a method of making a laser array having a planar active layer and a guide layer on a channeled substrate. In this method, a mesa having a plurality of corrugations on the surface is formed on the surface of the substrate, and then a first clad layer, an active layer, and a second clad layer are formed on the channel by liquid phase epitaxy. It includes the steps of sequentially depositing layers to form wider electrical contacts. Meltback of the convex portions of the corrugations during the initial process of deposition forms a channel between them together with the land portions and also retards the growth of the layer on the channel, thereby reducing the channeled surface. A flat layer can be formed on top.

[Brief explanation of the drawing]

FIG. 1 is a perspective view of a first embodiment of the phase-locked array of the present invention.

2 and 3 are cross-sectional views of second and third embodiments of the phase-locked array of the present invention.

4 to 6 are cross-sectional views of the substrate at various steps in the formation of a mesa with channels on its surface.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the drawings, identical elements in different embodiments of the invention are provided with the same reference numerals.

In FIG. 1, a laser array 10 comprises a body 12 of single crystal semiconductor material having regularly spaced parallel end faces 14 that reflect light at the laser wavelength and at least one of the end faces 14 from which light is directed. It is partially transparent so that it emits radiation.
The body 12 also has spaced substantially parallel sides 16 extending perpendicularly between the end faces 14.

Body 12 has spaced parallel first and second major surfaces 20, 22 extending perpendicularly between and perpendicular to end face 14 and side face 16, respectively. A mesa 23 having a surface 24 is formed on the first main surface 20 . A plurality of regularly spaced substantially parallel V-shaped channels 26 include:
Extending a distance from surface 24 into mesa 23 between end faces 14 . A first cladding layer 28 covers the substrate and mesa surfaces 20, 24, respectively, and fills the channel 26. A cavity region 30 is formed on the first cladding layer 28, consisting of a guide layer 32 overlying the first cladding layer 28 and an active layer 34 overlying the guiding layer 32. A second cladding layer 36 is formed over the cavity region 30 and a cap layer 38 is formed over the second cladding layer 36. Aperture 4 extending over channel 26
An electrically insulating layer 40 having an electrically insulating layer of 2 is formed on the cap layer 38. A large area electrical contact 44 is formed on the cap layer 38 and the electrically insulating layer 40 in the area of the aperture 42. A substrate electrical contact 46 is formed on the second major surface 22 of the substrate 18 .

The laser array 100 shown in FIG.
1 in that channel 102 extends a certain distance into channel 2 of array 10 in that channel 102 extends a certain distance into
It is different from 6.

Laser array 200 shown in FIG. 3 differs from array 10 in that V-shaped channels 202 extend a distance into the substantially planar major surface 204 of base 18. Laser array 200, shown in FIG. Array 200 includes cladding layers 28 and 36, guide layer 32, and active layer 3.
4 differs from arrays 10 and 100 in that the guide layer 32 and active layer 34 are thickest over the channel 202, with the guide layer 32 and active layer 34 varying in thickness laterally.

The laser array of this invention is formed of materials such as GaAs and AlGaAs with the necessary differences in reflection coefficients. Besides this, other group or group V elements such as Inp, Ga and As can also be used. Alloys that can be used for particular layers of the array are those in which the reflection coefficient 34 of the active layer 34 is greater than the reflection coefficient of the guide layer 32;
The reflection coefficient of the cladding layers 28 and 36 must be greater than that of the cladding layers 28 and 36.

The base 18 and the eleventh cladding layer 28 are of one conductivity type, the second cladding layer 3 and the cap layer 38 are of the opposite conductivity type. Cavity region 3
0, the positions of the guiding layer 32 and the active layer 34 can be exchanged with each other. The guide layer 32 is preferably located between the first cladding layer 28 and the active layer 34, in which case the guide layer 32 is located between the first cladding layer 28 and the active layer 34.
It has the same conductivity type as 8. In some cases,
The guide layer 32 includes an active layer 34 and a second cladding layer 36.
In this case, the guide layer 32
has the same conductivity type as second cladding layer 36.

The base 18 is preferably made of N-type GaAs with a first major surface 20 parallel to 100 crystallographic planes. The substrate may be deviated from this orientation, but 1
It is desirable to use the 00 plane. Channels should preferably be evenly spaced and generally about
At a depth of 1.5μm (microns) to 2.5μm, surface 2
The width at 0 is approximately 3.5 μm to 4.5 μm, and the typical center-to-center distance between channels is approximately 4 μm to 6 μm.
It is becoming. Center-to-center distances greater than those listed above are also valid, with the other dimensions varying accordingly. Channels are generally V-shaped with flat surfaces between them. Alternatively, the channel may have other shapes, such as a flat bottom channel as shown in FIG.

The first cladding layer 28 is typically N-type Al _Y
Consisting of Ga _1-r As, r is approximately 0.20 to 0.45,
Preferably it is about 0.25 to 0.35. This layer 28 is relatively thin over the land area between the channels, approximately 0.1μ
0.4 μm, typically about 0.25 μm, and preferably fills the channel and provides a flat surface of the first cladding layer 28 on which the next layer is deposited. is forming. Alternatively, the coverage of the first cladding layer 28 may be controlled such that the channel is not completely filled with the cladding layer 28, so that the surface of the first cladding layer 28 is curved.

Guide layer 32 is generally formed of N-type Al _x Ga _1-x As, with x between about 0.15 and 0.30, preferably between about 0.18 and 0.25. This layer is normally flat and its thickness is between about 0.3 μm and 0.6 μm, preferably about 0.4 μm. Typically about 0.3 μm on the channel, unless the layer is flat as shown in Figure 3.
The thickness is between 0.6 μm and about 0.1 μm to 0.4 μm on the land portion between the channels.

Active layer 34 is typically comprised of Al _y Ga _1-y As, where y is in the range of about 0.0 to 0.15, preferably about 0.03 to 0.12, and is preferably undoped. If active layer 34 is planar, it is typically about 0.05 μm to 0.12 μm thick. If it is formed on a curved surface, it is usually about 0.05μm to 0.12μm on the channel.
It is thinner over the flat land area between the channels, but is not zero.

The second cladding layer 36 is generally P-type Al _z Ga _1-z
z is between about 0.25 and 0.45, preferably between about 0.28 and 0.35. This layer generally has a thickness in the range of about 0.18 .mu.m to 1.5 .mu.m.

Cap layer 38 is generally comprised of P-type GaAs and is used to facilitate electrical ohmic contact between the underlying semiconductor material and the overlying metal contact. This layer is usually about 0.5μm to 1.0μm
The electrically insulating layer 40 preferably comprises silicon dioxide deposited on the cap layer 38 by thermal decomposition of a silicon-containing gas such as silane in oxygen or water vapor. Apertures 42 are drilled through electrically insulating layer 40 using standard photolithographic masking techniques and etching processes to form cap layer 3.
It is formed up to 8. A large area electrical contact 44 is then deposited onto the cap layer 38 exposed in the aperture 42. Large area electrical contact 4
Preferably, 4 is formed by sequentially evaporated titanium, platinum, and gold. Electrical contacts 46 for the base 18 are deposited on the main surface 22 of the base 18 by sequentially depositing and sintering tin and silver, followed by plating with a layer of nickel and a layer of gold. As described by Ladany et al. in U.S. Pat. No. 4,178,564, the emitting end face 14 of the array is typically made of Al ₂ O ₃ with a thickness of about 1/2 the wavelength of the laser light.
or covered by a layer of similar material. As described by Caplan et al. in U.S. Pat. No. 3,701,047, the opposite end face 14 is covered with a reflective surface of an electrically insulating material such as SiO ₂ coated with a layer of gold. . Alternatively, the mirror may be multilayer anti-beveled, as described by Ettenberg in US Pat. No. 4,092,659. These three US patents are incorporated herein by reference.

The laser array of this invention uses well-known photolithographic masking techniques and etching processes;
Channels are then formed by standard liquid phase epitaxy techniques as described by Lockwood et al. in U.S. Pat. No. 3,753,801, incorporated herein by reference, and further as described in U.S. Pat. Botez in the issue specification
It is constructed by depositing a layer on the basic surface containing the channel by the method described by Mr. Etching processes to form channels can include selective chemical etching or ion etching of surfaces with particular crystallographic orientations.
These techniques are well known in the art.

In addition to the method described above, it is also possible to first form a series of adjacent V-shaped channels, thereby forming a corrugated surface, as shown in FIGS. 4 and 5.
In FIG. 4, a GaAs substrate 300 is preferably formed on a major surface 302 of its 100 crystallographic planes.
It has a plurality of stripes 304 of an etch-resistant material such as SiO ₂ . The stripes are formed using standard photolithographic masking techniques and etching processes and are preferably aligned along the 011 crystallographic direction on the 100 orientation plane. A selective etch is then performed on the exposed surface of the substrate to form a V-shaped channel as shown in FIG. V-shaped channels 402 are formed by etching beneath the stripes 304 to the point that only a small portion of the original surface remains to support the stripes 304. The surface 302 outside the area of the stripes is also removed to form a new surface 404, leaving a mesa with a plurality of corrugations on the surface.

A substrate having a corrugated surface is inserted into a liquid phase epitaxy apparatus such as that described by Lockwood et al., supra, and contacted with a solution from which a first layer is deposited on channel 402 and surface 404. The sequence of events that occur early in the liquid phase epitaxy deposition process is highly dependent on the properties of the solution and the geometry of the substrate surface in contact with the solution. In the simplest case, a saturated solution of the element to be deposited and the element being a solvent for the basic material are brought into contact with a flat surface.
At this point, the substrate and solution are in equilibrium so that no deposition or meltback of the substrate occurs. When the substrate and solution combination is then cooled, the solution becomes supersaturated and deposition occurs.

However, if the surface is not flat and has a locally varying radius of curvature, the saturation of the contacting solution will also vary locally. If the solution is just saturated on a flat surface, the concave portions of the substrate surface will be supersaturated when viewed from the solution side, and the convex portions of the substrate surface will be unsaturated. Two effects therefore occur on the curved portion of the surface. Firstly, deposition occurs on the concave portions of the surface where the solution is supersaturated, and secondly, dissolution occurs on the convex portions of the surface where the solution is undersaturated.

These principles can be applied to a base 300 as shown in FIG.
When applied onto a corrugated surface, the convex portions of the corrugations, i.e. the tips of the protrusions forming the channels, undergo meltback and form flat land portions between the concave portions of the corrugations forming the channels. This meltback retards growth on newly formed land areas. Growth proceeds within channel 402;
After a predetermined period of time has elapsed, the surface of the adhered layer on the channel 402 and the surface of the land portion 502 become flat. Thereafter, growth occurs uniformly over the entire flat surface.

By carefully controlling the temperature of the solution, the degree of supercooling used, and the cooling rate, AlGaAs is deposited on the corrugated surface to form a flat layer on the channel.
It has been found that a cladding layer can be applied. In particular, the initial growth temperature is the temperature at which the solution equilibrates, generally 700°C to 800°C, preferably about
It was found that the temperature should be set at 760°C.
The temperature range at which the solution and base are cooled to low temperatures is between about 2°C and 10°C, preferably between 4°C and 5°C. The cooling rate is about 0.5°C to 5°C per minute, preferably 1°C. In this temperature range, the growth rate is significantly slower than the commonly used growth temperature range between 850°C and 950°C. Furthermore, convex meltback can be controlled or filling within the channels occurs more easily to form a flat surface. The higher the temperature, the
The thermal decomposition rate of the corrugated surface is higher due to the loss of arsenic.

The steps of the novel method for constructing a phased array of closely spaced lasers are to form a plurality of adjacent corrugations on a basic surface with a corrugation axis extending between the end faces. Generally, a solution containing the elements to be deposited is brought into contact with the source wafer as described by Lockwood et al. in U.S. Pat. No. 3,741,825, incorporated herein by reference. and is generated at the first temperature. Preferably, the solution and substrate are cooled separately through a temperature range that produces supersaturated growth conditions for concave and planar portions, and slightly undersaturated growth conditions for convex surface portions. The solution and the substrate are brought into contact by sliding the wafer into contact with the solution, thereby creating a partial meltback of the convexities or peaks of the corrugations, and meltback of the concavities or channels of the corrugations between the channels. Forms the land part. Deposition of the first cladding layer begins and continues until a flat surface of the deposited layer is formed over the protrusions and depressions of the original corrugations. The remaining semiconductor layers of the laser array are then deposited on this planar surface using standard liquid phase epitaxy techniques.

Operating the laser array under forward bias conditions causes current to flow through large area electrical contacts that extend laterally across the active channel. A lasing operation occurs on each channel of the active region, with a laser light beam propagating within both the active layer and the guide layer on each channel. The radiation from the individual oscillators does not use any lateral conductivity variations to confine the current to the portion of the active layer on the channel, as is generally required for this type of single oscillation. It was found that this was the basic lateral mode. The combination of uniform current from large area contacts combined with absorbing fundamentals in close proximity to the active and guiding layers on the land portion between the channels gives rise to oscillations only in the fundamental lateral modes. It is clear that this is sufficient. In this configuration, by placing the individual oscillators very close to each other, this is possible without the loss of excessive threshold current.

Coupling between oscillators on adjacent channels takes place by superposition of their instantaneous optical fields. This coupling is done with a 0°C phase shift between the oscillators, each of the coupled oscillators capable of oscillating in a fundamental lateral mode and producing a single output beam perpendicular to the end face. Alternatively, for an oscillator oscillating in the fundamental mode, the oscillator's lateral separation can produce a pair of symmetrical output lobes in the output beam separated by an angle of approximately 5° to 10°. Coupling can also be performed with a 180° phase shift between the two.

The invention will be illustrated by the following example. However, the invention is not limited to the detailed examples described below.

EXAMPLE A nine-element phase-locked laser array was constructed using the method of the invention. A mask was formed on the 100 surface of an N-type GaAs base wafer having groups of stripes of SiO ₂ containing 2 μm openings between 3 μm stripes. The longitudinal direction of the stripes is aligned along the 011 crystallographic direction of the substrate. Then the base is
It is etched in an etching solution consisting of H ₂ SO ₄ :8H ₂ O ₂ :8H ₂ O to form channels. The etching of the SiO ₂ stripes forms approximately triangular channels with a depth of approximately 2.2 μm and a center-to-center spacing of 5 μm. The material outside the channel area is also removed so that a triangular portion of the substrate material protrudes from the surface formed by the etching of the remaining portion of the substrate for each set.

Then, 3 grams of Ga in one bottle (container),
25 milligrams (mg) GaAs, 1.9 mg
The substrate is inserted into a multi-bin boat of the type described by Rozkuud et al. containing a solution prepared by mixing Al, 200 milligrams of Sn. The solution is equilibrated to the GaAs source wafer at 760°C. The growth solutions for the substrate and first coating layer are then individually cooled from the initial temperature of 760°C to about 4°C to 5°C at a rate of about 1°C per minute. The substrate is then contacted with the solution to deposit the first cladding layer. The initially triangular protrusions are meltbacked by about 1.3 .mu.m, forming a 0.9 .mu.m deep triangular channel with a flat land section between them. The deposition of the layers is then carried out in the following order: 0.25 μm thick N-type Al _0.3 Ga _0.7 As layer on the land part, 0.4 μm thick N-type Al _0.22
Guide layer of Ga _0.78 As, 0.06 μm thick Al _0.07 Ga _0.93
As active layer, 0.8μm thick P type Al _0.35 Ga _0.65 As
A second cladding layer of 0.3 μm thick and a P-type GaAs cap layer are deposited. An approximately 0.1 μm thick SiO ₂ insulating layer is deposited on the cap layer, and then 50 μm wide apertures for large area contacts are formed on the channels using standard photolithography and etching techniques. . Ti, Pt and Au are deposited on the oxide and cap layer by vacuum evaporation.
Electrical contacts to the substrate are formed by vacuum deposition of Ag and Sn followed by a sintering step. This substrate is then plated with Ni and further coated with Au.

The wafer is then cut into strips. First strip of paper
is coated with Al ₂ O ₃ with a thickness of about 0.27 μm, and the second
The surface is coated with six dielectric laminated reflective layers. The individual dice made from the strips are then assembled for testing.

The device is tested in pulsed mode using 100ns (nanosecond) pulses of continuous waves at a frequency of 1KHz. Individual devices exhibited threshold currents of 250 ma to 400 ma, peak pulse outputs up to 400 mW, and continuous wave outputs up to 80 mW. Several devices tested exhibited a two-rope far-field pattern consistent with a 180° phase shift motion. Other devices exhibited single lobe characteristics with 0° phase shift operation in pulsed mode operation. The properties of these far-field patterns were improved by increasing continuous wave power indicating that the coupling between the radiating elements increases with increasing drive level.