JPH06224523A - Laser array - Google Patents

Abstract

PURPOSE: To provide a multi-beam semiconductor laser which is capable of generating multiple beams provided with a different wavelength or different polarization. CONSTITUTION: A structural body on a right side, constituted of a contact point 25, a substrate, layers 11-14 and a contact point 26 forms a first QW laser provided with an active region in the layer 12. The structural body at a center part and on a left side formed by a contact point 27, layers 15-19 and a contact point 28 forms a second QW laser, provided with an active region in the layer 17. By selecting the active region layers 12 and 17 for providing different formation components, laser beams obtained are provided with different wavelengths.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION This invention relates to solid state lasers capable of delivering multiple radiation beams in close proximity.

[0002]

Solid state light sources are well known which have the inherent advantages when used as light emitters for electro-optical line modulators in printing equipment. For example, US Pat. No. 4,78
No. 6,917, the contents of which are incorporated herein by reference. Proposed here as a solution to the problem is to provide an array of closely spaced laser emitters that continue to operate uncoupled laser light emission, or by diagonal displacement of individual laser emitters, or , To form regions of optical isolation between the laser cavities of the array using impurity-induced disordering (IID). The environment of this solution is that the individual emitters of the beam have the same basic parameters, in particular:
It has the same wavelength and bias. This problem,
This is exacerbated when constraints are imposed such that adjacent lasers have different beam parameters, eg, adjacent beams have different wavelengths and / or different polarizations.

Among other things, the highly desired performance enhancement of this modulator is preferably in close proximity on the same wafer, but with a relative wavelength shift of a few nanometers. It can be achieved by manufacturing two sources. By using a dispersive element, the wavelength shift can be converted into a linear relative deviation of the spot in the image plane for the two sources, so that the angular displacement of the two beams in the plane results in a beam scanning mechanism Beyond that, this allows the two beams to be directed to different development stations, allowing easy multicolor xerography. Certain wavelength shifts can be achieved by changing the driving conditions of the semiconductor lasers used in the modulators described in the patents, which shifts are generally required to improve the performance of this device. Not large enough for convenient design of decentralized elements.

These also represent a need in the art for multiple beam sources in which the output beam can be split by wavelength and polarization splitting techniques. Such a light source
For example, it would include the following: (1) Multiple stacks with multiple wave outputs obtained by adjusting the configuration of the active (active) regions to have different wavelengths. Full laser structure (fu
ll laser structure), (2) US Patent Application No. 948,522 (1992) related to the present application.
22 September 2013), a transverse electric field with polarity switched by an intracavity loss mechanism.
lectric) (TE) and transverse magnetic
) Designing a semiconductor integrated circuit of a laser structure with a single active region quantum well (QW) capable of enabling both (TM) gain, or (3) as described in the above patent application. One of the wells outputs the TE bias, the other well outputs the TM bias, and the output of each bias type is an intracavity loss mechanism (intracavity locus).
2 or more QWs as selected by ssmechanism)
Monolithic design of a single active area laser structure including.

[0005]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a multi-beam multi-beam solid-state laser with different wavelengths or different polarisations on a common substrate and a method of making such a solid-state laser. .

[0006]

According to one aspect of the invention, two independently addressable lasers on the same substrate emit beams separated by a large controllable separation in wavelength. To do. The laser is of QW type and is composed of stacked epitaxially grown semiconductor layers having at least two laser active (active laser) regions. The stack of layers comprises a first active region in which one set (combination) of layers can be selectively activated (actuated) to emit a first beam,
Another set of layers is constructed and contacted with a second active region that can be selectively activated to emit a second beam.

According to one aspect of the method described in the present invention, a stack of both active layers is grown on a common substrate, and the resulting layers are first active area only. Of the second active region and only the second active region is configured and contacted in the second laser. The wavelength separation is thus determined during the growth of the epitaxial layer, and
It is a design parameter independent of the driving conditions of the device. An important advantage of the present invention resides in a structure in which both lasers are essentially similar, but emit different wavelengths due to the different underlying QWs.

The method described in this embodiment of the present invention selectively etches the epitaxial layer of the topmost laser to form the desired buried set of laser layers for forming another laser. It is to contact.

A further described method is to excavate one or more diffused or implanted regions through an epitaxial layer overlying 1
Contacting one or more embedded sets of laser layers.

In accordance with another aspect of the present invention, TE bias and TM are eliminated without having to rely on intracavity loss mechanisms.
Multiple lasers that can include any of the biases are:
It can be accomplished by methods similar to growing full laser structures on top of each other. US Patent Application No. 94
No. 8,528 (Filing date: September 22, 1992, Name:
MULTI-BEAM, ORTHOGONALLY-POLARIZED EMITTING MONOLIT
HIC QUANTUM WELL LASERS (diagonally biased multi-beam radiation monolithic QW laser) and US patent application No. 948,522 (filing date: September 2, 1992).
2nd, Name: POLARIZTION SWETCHABLE QUANTUM WELL LA
The QW laser manufacturing technology described in SER (biased switchable QW laser) is a stacked laser, one set produces a TE mode beam and the other set produces a TM mode beam. Generate,
Both sets may be included in a stacked laser as previously determined by forming multiple layers and making suitable contacts. In the described method,
Portions of the layer are selectively stripped from the stack to form one or more lower laser lasing structures, and other layer portions remain and the higher laser To form a structure that functions as. The top structures are accessed directly on the top surface, and the bottom structures are accessed by etching the top structure or structures to expose the bottom structure or structures. To be done. Individual laser stripes (la
ser stripes) is, for example, rib guide etching, I
It can be processed by any suitable technique, such as ID and the like. The spacing between the devices is chosen to be as narrow as possible as long as acceptable contact is possible and acceptable levels of crosstalk are created. Vertical spacing of QW lasers and delamination of selected layers may be useful in blocking electrical, optical and thermal effects that cause unwanted crosstalk.

As an alternative to the selective stripping technique, the diffused or implanted regions can be selectively embedded in a layered structure to form a plurality of emissive regions. To provide such desired selective access and contact.

Contacting to provide independent addressing may be performed in different ways. One way is
The use of a common, lower contact and holding a near TE polarized lowest threshold device on top of the stack. It will occur (raise) at a much lower current than the underlying structure, and will therefore be addressable without producing other biased light output. In the lower, higher threshold structure, everything above it has been removed and is addressed independently. Alternatively, an intermediate side contact point can be used beside each laser structure, then
The voltage may be applied independently between each structure. This embodiment is also essentially similar in its features to a QW type laser.

The following is a list of these different aspects.
(1) A laser array comprising: (a) a normal substrate having a semiconductor substrate portion; and (b) a semiconductor layer laminated on the substrate portion, the first layer being a first layer. A first set of layers capable of supporting a first laser radiation QW active layer generating a first beam of radiation having a parameter,
A second set of layers capable of supporting a second laser radiation QW active layer formed with or without the substrate portion to generate a second beam of radiation having a second parameter is the substrate. Formed with or without a substrate portion, (c) a first portion of the substrate for generating the first beam of radiation having a first parameter by the first set of layers. By means of contacting a first portion of the substrate to form a first QW laser having a first active layer, and (d) a second portion of the substrate by a set of the second layer. Means for contacting a second portion of the substrate to form a second QW laser having the second active layer for generating the second beam of radiation having a second parameter. A laser array characterized in that.

Further (2) is a laser array:
(A) a common substrate having a semiconductor substrate portion, and (b) a general planar stack of semiconductor layers on the substrate portion,
A first set of layers, the stack of first layers being capable of supporting a first laser radiation QW active layer that produces a first radiation beam having a first parameter, is provided with or on the substrate portion. A second set of layers capable of supporting a second laser radiation QW active layer formed with no substrate and generating a second radiation beam having a second parameter, with or without the substrate. Is formed, the first QW active region is at a different level than the second QW active region,
(C) a first QW laser having a first active layer for generating a first beam of radiation having a first parameter through a first portion of the substrate by the set of first layers. Means for connecting to form a first radiation beam having a second parameter by means of: (d) a second portion of the substrate by a second set of layers; Means for contacting to form a second QW laser having an active layer of, and (e) a region where the first and second connecting means are selectively diffused or implanted.
A laser array characterized in that.

Further, as a method of manufacturing these laser arrays, (3) a method of manufacturing a QW laser array, including:
(A) Prepare a normal substrate having a semiconductor substrate portion,
(B) forming a stack of semiconductor layers on the substrate portion, (c)
The stack is provided with first and second layers respectively capable of supporting first and second laser radiation QW active layers respectively producing first and second radiation beams respectively having first and second parameters. Two sets are treated to be formed with or without the substrate portion, and (d) a first portion of the substrate having a first parameter with the first set of layers. Contacting to form a first QW laser having the first active layer for generating the first beam of radiation, (e) setting a second portion of the substrate to the set of second layers Contacting to form a second QW laser having the second active layer for generating the second beam of radiation having a second parameter,
Examples thereof include a method for manufacturing a laser array, which includes each step.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A feature of the present invention is that a common body comprising a semiconductor substrate having a top layer on which a plurality of layers capable of holding active regions that can be lased to emit radiation beams of different parameters are epitaxially grown. Body). The structure then comprises a first body portion comprising a first active region emitting a first beam having a certain parameter and a second body portion emitting a second beam having a different parameter. Formed and contacted to include two active regions.
The most important aspect of the present invention is that among the stacked layers, one emits a first beam when properly contacted and another emits a second beam when properly contacted. It is a structure having a radiating stack. The result is a QW type laser array with multiple beams with different characteristics. Often, but not limiting of the invention, multiple beams are emitted from adjacent active regions of a common, often with only two types of beams being emitted, for example: Beams at two different wavelengths or at different polarizations, but again, the invention is not so limited and can produce three or more different beams.

The proposed structure utilizes two stacked sets of epitaxially grown layers, each set containing an active region, however, after growth, the first set of regions is: For contact purposes, a second set of layers including the second active area is etched. In this method, the lasing regions of each emitter are placed in different active regions with different wavelengths, independent of their driving conditions. The two lasers thus formed use two separate layers for their active regions and thus can be designed to have separate wavelengths independent of the associated driving conditions. This technique and concept can be independently utilized as the particular type of laser device design used to implement the illuminator. And for this reason,
These can be included in any past, present, and future QW type device design suitable for use as an electro-optical linear modulator.

FIG. 1 shows an example of a suitable set of epitaxial layers (scale not taken into account) for realizing a dual laser device, and FIG. 2 shows the subsequent two sets after patterning. 2 shows the same set of layers, each of which can be operated in an independent way 2
Contacts are added to provide two adjacent lasers.

FIG. 1 also illustrates a method of constructing two separate QW laser structures in a stacking manner on the same substrate, in accordance with the present invention. One common substrate 2
0 is prepared. In this embodiment, the substrate 20 is
It includes a GaAs N ⁺ substrate 10. Nine epitaxial semiconductor layers denoted by 11 to 19 are grown thereon by a general-purpose means. Many of the known techniques mentioned in the referenced patents and the related patents above may be applied, and these descriptions are also incorporated by reference. One example, which should not be considered limiting, uses known MO-CVD growth techniques. In this example and the following, binary, ternary, and quaternary III-V semiconductor composites are used. Because they are found in the state of the art to produce the most effective laser diodes in the visible and near-visible range.
However, the invention is not limited to such composites, but can be applied to any chemical system in which the layers of the stack can be used to fabricate a laser actuated active region.

The epitaxial layers 11 to 19 have the following compositions in this embodiment: Layer number Composition Function 11 N-type AlGaAs cladding layer (coating layer) 12 AlGaAs First active layer 13 P-type AlGaAs Clad layer 14 P-type GaAs contact (contact) layer 15 N-type GaAs contact layer 16 N-type AlGaAs clad layer 17 AlGaAs second active layer 18 P-type AlGaAs clad layer 19 P-type GaAs top (cap) layer The thicknesses of the various layers will be apparent to those skilled in the art given the function given. Typically, the QW type active layers 12 and 17 will be thinner than the remaining layers. From the description of FIG. 1, the substrate 10 forms a first potential quantum well (QW) laser with the first set of layers 11-14 and the second set of layers 15-19 has a second potential. To form a QW laser.

Independently addressable by further introducing impurities in selected regions in the substrate 20 and / or by other known isolation techniques and by providing suitable contacts. A parallel, array of QW lasers is obtained. The essential similarity of the two lasers simplifies the subsequent process. By using different sets of layers 11-19, denoted by reference numerals 22 and 23, the laser is capable of producing a radiation beam with different parameters. In this embodiment, the first and second lasers are configured to emit radiation beams of first and second wavelengths, respectively.

The preferred embodiment for avoiding cross-coupling and achieving the desired contact is:
Physically form the structure of FIG. 1 to obtain an independent laser. This is preferably done by using a general purpose masking technique and a general purpose etchant to produce the semiconductor structure of Figure 2 (without contact metals 25-28). Known etch stop layers can be used to limit the etching to the levels indicated. For example, GaIn between layers 14 and 15
By using a thin layer of P (not shown), H ₂ O ₂ / H ₃ P
Such etchant O ₄ is etched all the layers 15 to 19 are superimposed but stops at GaInP layer. Similarly,
By providing a GaInP thin layer (not shown) between layers 15 and 16, an etchant such as H ₂ O ₂ / H ₃ PO ₄ etches all of the overlaid layers 15-19, but GaInP does not.
Do not etch the P layer. Instead, AlGaInP
However, for an etching solution such as 5H ₂ SO ₄ : H ₂ O ₂ : H ₂ O,
It can be used as an etching stop layer.

Further, the etch stop layer between layers 15 and 16 can be omitted by using HCl or HBr as the etchant, which etches AlGaAs down to GaAs. Other known methods can also be used to limit the etching to the layer shown in FIG.

Following the etching process, suitable contacts, such as Cr-Au or Ti-Pt-Au metallization, can be provided to form the contacts of the illustrated layers 25-28. .

The structure on the right formed by the contact 25, the substrate 10, the layers 11-14 and the contact 26 forms a first QW laser having an active region in the layer 12. Contact 2
The central and left side structure formed by 7, layers 15-19 and contact 28 forms a second QW laser with an active region in layer 17. By choosing the active region layers 12 and 17 to have different constituents, the resulting laser beams will have different wavelengths. Not shown in FIG. 2 are conventional, general purpose, reflective edges and other structures (such as stripped confined regions) for preventing lateral diffusion of carrier particles and photons that amplify the laser action. It is the body. All conventional techniques can be readily applied to the manufactured discrete QW laser structure.

As described, when a suitable voltage is applied between contacts 25 and 26, current will flow only through the active region 12 of the first layer on the right side. Similarly, when a suitable voltage is applied between contacts 27 and 28, current will only flow through the central second region active region 17. Therefore, both lasers are addressed independently. The first and second lasers
By using a separation layer for their active regions, they can simply be designed to have separate wavelengths independent of their associated operating conditions. Further, the preferred stacked layers containing layers for making contacts and the semiconductor types used in these QW structures provide unique P-N-P structures that provide good electrical contacts. To manufacture.

FIG. 3 depicts a laser array 30 that has been put into practice, in which there are three different wavelengths λ _1
3 independently addressable QWs emitting at ~ λ ₃
It includes lasers 31-33. Figure 3 shows the final product. The configuration is similar to that of Figure 1, but in this case includes three active layers represented by oval shapes. The arrangement of the layers on the top of the GaAs substrate is from bottom to top (not shown in FIG. 3) as follows: Layer number Composition Function 1 N-type AlGaAs cladding layer (coating layer) 2 AlGaAs First active layer 3 P-type AlGaAs clad layer 4 P-type GaAs contact layer 5 N-type GaAs contact layer 6 N-type AlGaAs clad layer 7 InGaAs Second active layer 8 P-type AlGaAs clad layer 9 P- Other variations in the order of type GaAs contact layer 10 P-type AlGaAs cladding layer 11 AlGaAs third active layer 12 N-type AlGaAs cladding layer 13 N-type GaAs cap layer are also possible.

As in the embodiment of FIG. 1, the first active layer 35 on the left is at the lowest level, the second active layer 36 on the right is at the intermediate level, and the third active layer 37 in the center is It was the highest level. A common contact 38 is disposed on the bottom and metal strips 39-40 are disposed on each of the active areas, respectively, to serve as a second contact. Each laser can be independently addressed by a voltage applied between strips 39-41 and common contact 38. The vertical spacing of the two structures is preferably determined by the thickness of the epitaxial layer, and the lateral spacing is chosen to be as small as possible in contact and crosstalk limits. In this case, the lateral spacing between the active areas is about 55 um, with the lowest part 35 and the last part 3
The spacing between the 7 active areas was approximately 5 um.

FIGS. 4A to 4C show a laser 31.
~ 33 show the respective power output vs. input current characteristics and other operating characteristics including wavelength.

The present invention is not limited to multiple lasers on a common structure producing beams of different wavelengths. Another parameter is also this theory of stacking multiple active layer (s) with separate contacts and means for separately contacting and driving to selectively generate multiple beam (s). Can be changed by using. In the embodiment of FIGS. 5 and 6, beams with different polarizations are generated.

The above identified relevant cases, US Patent Application Nos. 948,524 and 948,522.
In the publication, it is described that the structure produces a beam of transverse electrical (TE) and transverse magnetic (TM) polarization. Several techniques for making QW lasers that generate either TE or TM polarized beams have also been described. For example, if the active layer has a single Q
When the light hole and the heavy hole bands that include W and are generated at the same time in the energy decay state are generated at the same time, the emission bias is the threshold carrier density and temperature, facet reflection. Determined by other factors such as gender. The threshold carrier density can be altered by selective transfer defects in the laser structure. Also, in the case of certain structures (composites) described in the two QWs and the related cases referred to, there is no strain and the compressively strained material oscillates normally in the TE mode. A material strained with a certain tension can oscillate even in the TM mode.

These and similar teachings are for TE mode,
This is incorporated into the present invention by using an active region that oscillates in each of the TM modes to produce a bi-polarized beam from an independently addressable laser structure on a common substrate.

FIG. 5 is a schematic end view of one form of structure that provides a dual polarized laser having a common substrate contact. The structure of the substrate, semiconductor layer and contact metal structure is shown in this figure. Laser 50 on the left
Uses a substrate 51 and a set of left layers including a compressively strained QW active layer 52, a bottom contact 53 and a top contact 54 to produce TE polarized radiation at a wavelength of about 650 nm. . The laser 56 on the right side includes a substrate 51, a QW active layer 57, a bottom contact 53 and a top contact 5.
8 is used to produce a TM polarized beam at approximately the same wavelength. Even if the TM active layer 57 is also located on the left laser 50, it requires a higher threshold current than the active layer 52. As a result, the left laser 50 may oscillate in TE mode before the current reaches the threshold required to cause TM mode oscillation. The formation by etching of the laser 56 on the right side removed the TE active layer 52 on the right side. Al
The GaInP layer is located on the side of both active layers,
Functions as a limiting layer.

This embodiment is also shown to use etch stop layers 59 and 60 to prevent overetching.

FIG. 6 shows a dual polarization laser,
7 illustrates a similar embodiment providing a laser with independent, non-shared contacts. For similar components, see FIG.
The same reference numbers are used. The main change is the addition of an N-type layer 62 to the left laser 50,
The N-type layer 62 functions as a contact (dot) layer for the uppermost metal contact 63. Thus, the two contacts for the left laser 50 are 54 and 63, while the right laser 56 leaves 58 and 53. in this case,
Since there is no etch stop layer, selective etching is obtained, for example, by using an etchant that does not etch the aluminum-free layer 62, HCl: H ₂ O or HBr: H ₂ O, for the aluminum-containing layer. . Thus, layer 62 also functions as an etch stop element.

As a result, in a narrow interval, TE
A structure is realized for a monolithic laser source with independently addressable multiple (multi) beam outputs of either T and TM polarization. The structure does not require an intracavity defect mechanism that compromises the gain of one of the lasers as did monolithic structures, as described above.

In the preferred embodiment, each active layer is at a different height above the substrate. Preferably, the multi-layer stack is for compressive strained / unstrained quantum well structures or for tensile strained TM polarization as described in the relevant case referenced. To use. Preferably, the lower structure is
It is shaped to fit under more, higher threshold TM polarization, so that when a voltage is applied, only the vertically higher and lower threshold lasers are operated in TM mode. To achieve TM operation, the higher laser is either removed or a separate lateral contact is added and the laser is operated.

Although chemical etching techniques were used in these described embodiments to form layered structures in different sets with different active layers, the present invention is limited to this formation technique. is not. Ion milling and similar dry etching techniques well known in the art that rely on ion or electron bombardment to etch unmasked substrate regions may alternatively be used. Similarly, known precision machining techniques, such as a gas etching process selectively activated by a computer controlled laser beam, may also be used. As a further alternative, an IC-based precision manufacturing process of the type described in IEEE Trans. Electron Devices Vol. 35, No. 6, pp. 724-730 (1988) may be used. These processes commonly use photolithographic techniques or computer control to allow precise formation and contact of various substrate areas. For this reason, the method according to the invention is precisely aligned with one another and is suitable for improving the resolution of printing, for example by multi-station, multi-color ROS or linear modular,
It solves the problem for manufacturing multi-wavelength or multi-polarized laser sources.

The vertical spacing of adjacent laser structures is preferably determined by the thickness of the epitaxial layer, and the lateral spacing is the contact and crosstalk.
It is selected as small as possible as long as it is compatible with the restriction of.

FIGS. 7 and 8 show a further embodiment of the invention employed in forming by selective removal of layers to provide a stepped structure. Various layers
In order to facilitate the understanding of the following new themes, which have the following meanings, they are labeled in the figures; clad is an abbreviation for clad layer; SCH stands for separation (confinement). ) It is a captain character of a heteromorphic structure, and is well known and thin.
It is meant to add a high QW index index optical confinement layer on both sides of the QW layer to provide a waveguiding layer for the generated radiation. This is the case, for example, with two 0.14 um AlGaI flanks of the QW layers 52 and 56 in FIG.
Corresponds to the nP layer.

The barrier reduction layer is a known layer added to reduce the series resistance of the laser diode.

The embodiment of FIG. 7 differs from the geometry described above in that the lower diode structure 70 on the left is in contact with the bottom contact 72, with the N layer 71 at the bottom and the P ⁺ layer 73 at the top. The difference is that it has the usual arrangement of layers in contact with the contacts 74 at. The laser diode 75 on the right has the same P ⁺
Layer 73 and contact 74 are in contact and N ⁺ layer 76 and its contact 77 at the top have four contacts in the opposite arrangement.
The different structures of the SCH / QW layers 78, 79 result in different wavelength outputs. Furthermore, an etch stop layer is not needed in this embodiment.

In the embodiment of FIG. 8, there are two conventional laser diode arrays. The laser 70 on the left (using the same numbers) is the same as in FIG. 7, but the laser 81 on the right includes an additional N ⁺ layer 82 with a contact 83 and a top P ⁺ layer 84 and its. Cooperate with contact layer 85 to form contact pairs.

The embodiment of FIG. 8 was carried out with the following results. The AlGaAs laser on the left produces a 837 nm laser. Its voltage output / current characteristic is 200
It had a threshold at A / cm ² and an efficiency (curve slope) of 25.8% / facet. AlGaIn on the left
The P-laser emits a laser at 690 nm and emits 260
It had a threshold of A / cm ² and an efficiency of 20.6% / facet.

9 and 10 show two additional dual beam embodiments used by forming by diffused or implanted regions to form a planar structure. Again, the various layers are labeled to aid understanding. Two lasers are formed which emit in the oval areas indicated by 87 and 88.
As can be observed, in the example of Figure 9, the layers are all P-type. The desired NP junctions and contacts are formed by N-type selective diffusion 89, 90 forming impurities such as silicon. The diffusion 89 on the left side is the upper active layer 9
1 and the lower active layer 92 and penetrates to reach the P-clad layer 93 below the AlGaAs laser. The diffusion 90 on the right is the other P of the AlGaAs laser.
Penetrating to reach the cladding layer 94. A on the left
The two contacts of the lGaAs laser are shown at 95,96. The two contacts of the AlGaInP laser on the right side are
Shown at 97, 98. Areas 100 and 101 labeled "etched" are the top P ⁺
It has been removed to prevent unwanted lateral current flow through the GaAs layer 102. Alternatively, the diffusion or implant can be replaced to convert the material in regions 100, 101 to a semi-insulating material to prevent unwanted lateral current flow.

In the variant shown in FIG. 10, similar components are given the same reference numbers and N ⁺
A cladding layer 104 is provided for the AlGaAs laser on the left, and a single N-type diffusion 105 is the only active layer 9
Required to form a common contact 106 for both 1,92.

The general structure described above is similarly adapted to IR and visible light generating lasers. As the bonded structure grows, the total height of the epitaxial layers should be as low as possible to keep the quality of the film as high as possible throughout the structure. The total thickness can be maintained by minimizing the thickness of the cladding layer.

The technique of the present invention for application in QW lasers is extended by combining the TM and TE polarized outputs independent at one wavelength with the TM and TE polarized outputs at another wavelength. You can. The four beams produced can be split by a dichroic technique and sent to four xerographic stations for four-color printing. The structure of FIGS. 5 and 6 is 640 to 650 nm.
Designed for radiation in. However, the emission wavelength can be changed by adjusting the strain, thickness and Al content of the quantum well. Thus, four QW laser structures emitting at two wavelengths and two polarizations are
Etched back to expose all four structures and then contacted using the technique shown in FIG.
It can be easily grown in a stack, and the stack is. Alternatively, two different structures of the type in FIG. 5 are grown, one with TE and T at λ ₁ .
It emits M-polarized light, the other emits TE and TM polarized light at λ ₂ . These two structures are then brought into close proximity to each other by hybrid technology.

The advantages of the present invention are as follows: (1) Addressing independently at different wavelengths and / or at different wavelengths, TE and TM polarized, by using a QW laser structure. Possible monolithic sources emitting multiple beams are possible; (2) no intracavity loss mechanism is required to select the output polarization; (3) vertical displacement and etch removal or selective diffusion. / The implant provides isolation for the thermal, optical and electrical components of crosstalk; (4) This concept combines wavelength and polarization slitting into a single formation or monolithic / hybridization Extended to allow four-color printing; (5) The final products are essentially similar and independently addressed. It is possible, become a QW laser of narrow interval,
It is easily manufactured and can use well-established known techniques.

Methods for growing QW laser structures described, or similar to those described, are well known in the art, and they are numerous publications as well as the references cited herein. It can also be found in published papers. Examples of these papers include, for example, Apply of Physical Letter (Appl. Phys. Lett.) Vol. 60, No. 16, pp. 1927 to 1929 (published on April 20, 1992) and references cited therein; And S
PIE Vol. 1582 "Integrated Optoelectronics for Communication and Processing" Vol. 194-2
Page 05 (published in 1991) and references cited therein are included.

It will be understood that the invention is furthermore not limited to the particular combination of substances given by way of example in this case, however. All of the case combinations described are also included, along with the referenced prior art suitable for making QW lasers. A particularly useful combination is AlGaAs or InGaAs or AlGa
InP as an active layer, AlGaAs or AlGa
The cladding layer containing InP is located on the side surface.

[Brief description of drawings]

1 shows a cross-sectional view of a suitable intermediate structure for forming the dual laser device of the present invention.

FIG. 2 shows a cross-sectional view of an intermediate structure for providing a dual beam laser that produces different wavelengths.

FIG. 3 shows a perspective view of a triple-beam QW laser product emitting three different wavelengths, similar to the laser of FIG.

4 (A) to FIG. 4 (C) show characteristics of output power-input current characteristics of each of the lasers shown in FIG.

FIG. 5 shows a cross-sectional view of one embodiment of a structure with two polarized lasers in contact with a conventional substrate.

FIG. 6 shows a cross-sectional view of a similar embodiment with two polarized lasers, the lasers having independent, non-distributed contacts.

FIG. 7 shows a cross-sectional view of an embodiment employed to form a layer by selective stripping to provide a stepped structure.

FIG. 8 shows a cross-sectional view of an embodiment of the invention in which there are two conventional laser diode arrays.

FIG. 9: Generates two additional beams using diffused areas to form a planar structure, all layers P
Figure 3 shows a cross-sectional view of a mold embodiment.

FIG. 10 shows a cross-sectional view of an embodiment in which an N ⁺ cladding layer was placed for the left laser and a single N-type diffusion was used for both normal contacts of the active layers.

Continued Front Page (72) Inventor Gregory Jay. Kovacs, California, USA 94086 Sunnyvale Lily Avenue 1053 (72) Inventor David Pee. Baua United States California 95014 Spring Court Cupertino Fall Creek 11577

Claims (1)

[Claims] 1. A laser array comprising: (a) a common substrate having a semiconductor substrate portion, and (b) a semiconductor layer laminated on the substrate portion, wherein the first and second parameters are provided, respectively. A first and a second set of layers capable of supporting a first and a second laser radiation QW active layer, respectively, for generating a first and a second radiation beam, respectively, together with the substrate portion or the substrate. (C) the first active for generating the first radiation beam having a first parameter by means of the stacked semiconductor layers formed without a portion; and (c) the set of first layers. Means for contacting a first portion of the substrate to form a first QW laser having a layer, and (d) by the set of the second layer, the second radiation beam having a second parameter. The second active to generate Laser arrays, characterized in that it comprises a means for contacting a second portion of the base body so as to form a second QW lasers with. (C) a first QW having a first portion of the substrate with the first active layer for generating the first beam of radiation having a first parameter with the first set of layers. Means for connecting to form a laser; and (d) a second portion of the substrate for generating the second radiation beam having a second parameter by the second set of layers. Means for connecting to form a second QW laser having a second active layer;