OPTOELECTRONIC DEVICE WITH ENHANCED LATERAL LEAKAGE OF HIGH ORDER TRANSVERSE OPTICAL MODES INTO ALLOY-INTERMIXED REGIONS AND METHOD OF MAKING SAME

20170317472 · 2017-11-02

Assignee

Inventors

Cpc classification

International classification

Abstract

Optoelectronic device undergoes selective chemical transformation like alloy compositional intermixing forming a non-transformed core region and an adjacent to it periphery where transformation has occurred. Activated by selective implantation or diffusion of impurities like Zinc or Silicon, implantation or diffusion of point defects, or laser annealing, transformation results in a change of the refractive index such that the vertical profile of the refractive index at the periphery is distinct from that in the core. Therefore the optical modes of the core are no longer orthogonal to the modes of the periphery, are optically coupled to them and exhibit lateral leakage losses to the periphery. High order transverse optical modes associated to the same vertical optical mode have higher lateral leakage losses to the periphery than the fundamental transverse optical mode, thus supporting single transverse mode operation of the device. This approach applies to single transverse mode vertical cavity surface emitting lasers, edge-emitting lasers and coherently coupled arrays of such devices.

Claims

1. An optoelectronic device comprising a) a multilayer structure; b) a core region in the lateral plane; wherein said core region has a first vertical profile of the refractive index, wherein a set of optical modes having a first vertical profile of the optical field are formed in said core region, wherein said set of optical modes contains a fundamental transverse optical mode and at least one high-order transverse optical mode; and c) a periphery region in the lateral plane, wherein said periphery region is an adjacent region contiguous in the lateral plane to said core region, wherein said periphery region further comprises chemically transformed region, wherein said chemically transformed region is bounded by a bottom boundary and a top boundary, wherein said chemically transformed region is formed by a process of selective chemical transformation, wherein said process of selective chemical transformation changes the chemical profile in the vertical direction, wherein said selective chemical transformation in said chemically transformed region results in a change of the refractive index profile in said chemically transformed region such that said periphery region has a second vertical profile of the refractive index distinct from said first vertical profile of the refractive index, wherein a continuum spectrum of optical modes having at least one second vertical profile of the optical field is formed in said periphery region, and wherein at least one high-order transverse optical mode formed in said core region has a wavelength at resonance with the wavelength of at least one mode of said continuum spectrum of optical modes in said periphery region, and wherein said first vertical profile of the optical field and said second vertical profile of the optical field have an non-zero overlap integral, and wherein the lateral leakage losses of said at least one high-order transverse optical mode formed in said core region is larger than the lateral leakage losses of said fundamental transverse optical mode formed in said core region.

2. The optoelectronic device of claim 1, wherein the sum of said overlap integrals of said optical mode having a first vertical profile of the optical field with all said optical modes having second vertical profiles of the optical field exceeds ten percent.

3. The optoelectronic device of claim 2, wherein the sum of said overlap integral of said optical mode having a first vertical profile with all said optical modes having second vertical profiles exceeds twenty percent.

4. The optoelectronic device of claim 2, wherein the overlap integral of said optical mode having a first vertical profile with at least one said optical mode having second vertical profile exceeds ten percent.

5. The optoelectronic device of claim 4, wherein the overlap integral of said optical mode having a first vertical profile with at least one said optical mode having second vertical profile exceeds twenty percent.

6. The optoelectronic device of claim 1, wherein all high-order transverse optical modes of said set of optical modes having a first vertical profile of the optical field in said core region have larger lateral leakage losses than said fundamental transverse optical mode of said set of optical modes having a first vertical profile of the optical field in said core region due to the resonant coupling to said continuum spectrum of optical modes having said at least one second vertical profile of the optical field in said periphery region.

7. The optoelectronic device of claim 1, wherein said core region further comprises an active medium.

8. The optoelectronic device of claim 7, further comprising d) a means of generating non-equilibrium carriers in said active medium.

9. The optoelectronic device of claim 6, wherein said optoelectronic device operates as a single transverse mode optoelectronic device.

10. The optoelectronic device of claim 8, wherein said means of generating of non-equilibrium carriers are selected from the group consisting of: (i) current injection, (ii) photoexcitation, (iii) electron beam excitation.

11. The optoelectronic device of claim 1, wherein said process of selective chemical transformation is carried out by a means selected from the group of means consisting of: i) selective implantation of impurities, ii) selective diffusion of impurities, iii) selective implantation of point defects, iv) selective diffusion of point defects, v) selective laser annealing, and vi) any combination of i) through v).

12. The optoelectronic device of claim 11, wherein said process of selective chemical transformation is selected from the group of processes consisting of: (i) alloy compositional intermixing, (ii) alloy phase separation, and (iii) a combination of (i) and (ii).

13. The optoelectronic device of claim 11, wherein said diffusion of impurities is diffusion of at least one type of n-type impurity.

14. The optoelectronic device of claim 13, wherein said at least on type of n-type impurity is Silicon.

15. The optoelectronic device of claim 11, wherein said diffusion of impurities is diffusion of at least one type of p-type impurity.

16. The optoelectronic device of claim 15, wherein said at least one type of p-type impurity is Zinc.

17. The optoelectronic device of claim 1, wherein said multilayer structure further comprises (A) a bottom reflector, (B) a first cavity adjacent to said bottom reflector, (C) a top reflector adjacent to said cavity on the side opposite to said bottom reflector:

18. The optoelectronic device of claim 17, wherein said bottom reflector and said top reflector are selected from the group consisting of (A) a multilayer interference reflector, and (B) an evanescent reflector.

19. The optoelectronic device of claim 7, further comprising a first p-n junction, wherein said active medium is located within said first p-n junction.

20. The optoelectronic device of claim 19, further comprising a second p-n junction, wherein said second p-n junction is located in said periphery region at said bottom boundary of said chemically transformed region.

21. The optoelectronic device of claim 20, wherein the opening voltage of said second p-n junction exceed the opening voltage of said first p-n junction by at least zero point one Volt (0.1 V).

22. The optoelectronic device of claim 20 further comprising a barrier region, wherein said barrier region is located such that said second p-n junction is located within said barrier region, wherein said barrier region has at least one layer having an energy bandgap exceeding the energy bandgap of the rest of said multilayer structure.

23. The optoelectronic device of claim 18, wherein said multilayer structure is formed of materials selected from the group consisting of: (i) Gallium Arsenide, (ii) Aluminum Arsenide, (iii) an alloy of Gallium Arsenide and Aluminum Arsenide; and wherein said barrier region further comprises at least one layer of an alloy of Gallium Arsenide and Aluminum Arsenide with molar fraction of Aluminum in the range between twenty and forty-five percent.

24. The optoelectronic device of claim 1, further comprising at least one oxide-confined aperture.

25. The optoelectronic device of claim 7, wherein said active medium is selected from a group consisting of: i) bulk material; ii) single or multiple quantum well; iii) single or multiple sheet of quantum wires; iv) single of multiple sheet of quantum dots; v) any combination of i) through iv).

26. The optoelectronic device of claim 1, wherein the materials used for fabrication of said optoelectronic device are selected from the group consisting of: (i) III-V materials; (ii) III-N materials; (iii) II-VI materials; (iv) group IV materials; and (v) any combination of (i) through (iv).

27. The optoelectronic device of claim 8, wherein said optoelectronic device is selected from the group consisting of: (i) light-emitting diode; (ii) edge-emitting laser diode; (iii) surface-emitting laser diode; (iv) tilted wave laser diode; (v) tilted cavity laser diode; (vi) semiconductor disc laser; (vii) passive cavity surface-emitting laser; (viii) single photon emitter; (ix) emitter of entangled photons; (x) semiconductor gain chip.

28. The optoelectronic device of claim 17 further comprising (D) a second cavity located within said top reflector.

29. The optoelectronic device of claim 28, wherein said bottom boundary of said chemically transformed region is located within said top reflector between said first cavity and said second cavity, wherein said process of said selective chemical transformation is selective alloy compositional intermixing.

30. The optoelectronic device of claim 29, wherein said vertical optical mode in said periphery region having second vertical profile of the optical field, wherein said second vertical profile of the optical field has the intensity maximum within said second cavity, vanishes due to said alloy compositional intermixing in said chemically transformed region.

31. The optoelectronic device of claim 30, wherein said set of optical modes having said first vertical profile of the optical field in said core region have a vertical profile of the optical field having the intensity maximum in the first cavity, wherein said continuum spectrum of optical modes having at least one second vertical profile of the optical field is formed in said periphery region, and wherein said at least one second vertical profile of the optical field has the intensity maximum in said first cavity, and wherein at least one high-order transverse optical mode having said first vertical profile of the optical field formed in said core region has a wavelength at resonance with the wavelength of at least one mode of said continuum spectrum of optical modes having said second vertical profile in said periphery region having intensity maximum in said first cavity, wherein said first vertical profile of the optical field and said second vertical profile of the optical field have a non-zero overlap integral, and wherein said non-zero overlap integral exceeds ten percent.

32. The optoelectronic device of claim 31, further comprising an active medium.

33. The optoelectronic device of claim 32, wherein said active medium is located within said first cavity, wherein said optoelectronic device is a vertical cavity surface emitting laser.

34. The optoelectronic device of claim 32, wherein said active medium is located within said top reflector, wherein said optoelectronic device is a passive cavity surface emitting laser.

35. The optoelectronic device of claim 34, wherein said active medium is located between a plane of said bottom boundary of said chemically transformed region and said second cavity.

36. The optoelectronic device of claim 35, wherein said bottom reflector is n-doped, wherein said first cavity is n-doped, wherein a part of said top reflector located between said first cavity and said active medium is n-doped, wherein said part of said top reflector above said active medium is p-doped, and wherein said alloy compositional intermixing is carried out by means of diffusion of a p-type impurity, and wherein said p-type impurity is Zinc.

37. The optoelectronic device of claim 17 grown epitaxially on a substrate.

38. An array of optoelectronic devices comprising at least two optoelectronic devices according to claim 1.

39. The array of optoelectronic devices of claim 38, wherein the optical fields of said at least two optoelectronic devices of said array of optoelectronic devices are coherently optically coupled with each other.

40. The array of optoelectronic devices of claim 38, wherein said array is employed for the steering of a laser beam.

41. The array of optoelectronic devices of claim 38, wherein all optoelectronic devices forming said array are positioned on a single wafer.

42. The array of optoelectronic devices of claim 38, wherein said at least two optoelectronic devices forming said array are positioned on two different wafers.

43. The array of optoelectronic devices of claim 38, wherein said array of optoelectronic devices is further positioned in an external cavity.

44. The array of optoelectronic devices of claim 38, wherein said array of optoelectronic devices is a source of the primary light for a frequency conversion system.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0032] FIG. 1A. Schematic cross-section of a prior art vertical cavity surface-emitting laser (VCSEL) with an oxide-confined aperture.

[0033] FIG. 1B. Schematic dispersion curves of the optical modes in the planar multilayer structure mimicking the VCSEL of FIG. 1A revealing a spectral interval in which VCSEL modes cannot leak from the core to the periphery.

[0034] FIG. 2A. Schematic cross-section of a prior art vertical cavity surface emitting laser (VCSEL) wherein the current flow is controlled by proton bombardment, wherein proton bombardment determines the core and the periphery of the structure.

[0035] FIG. 2B. Schematic dispersion curves of the VCSEL optical modes in planar multilayer structures mimicking the VCSEL of FIG. 2A showing a spectral range of the modes which cannot leak from the core to the periphery.

[0036] FIG. 3A. Schematic cross-section of a planar multilayer structure mimicking a vertical cavity surface emitting laser.

[0037] FIG. 3B. Schematic dispersion curves of several optical modes of the multilayer structure of FIG. 3A.

[0038] FIG. 4A. Schematic cross-section of a vertical cavity surface emitting laser (VCSEL) with engineered oxide apertures promoting lateral leakage of the optical modes from the core to the periphery.

[0039] FIG. 4B. Schematic dispersion curves of the VCSEL optical mode in the core and of a tilted optical mode of the periphery of the planar multilayer structure mimicking the core and the periphery of the VCSEL of FIG. 4A, respectively. Leakage of the optical modes from the core to the periphery is possible.

[0040] FIG. 5A shows the optical power reflectance spectrum at normal incidence of the vertical cavity surface emitting laser (VCSEL) structure of FIG. 4A in the core region revealing a VCSEL dip at 850 nm.

[0041] FIG. 5B shows the optical power reflectance spectrum at normal incidence of the VCSEL structure of FIG. 4A calculated for the oxidized periphery region.

[0042] FIG. 5C shows the optical power reflectance spectrum of the VCSEL structure of FIG. 4A calculated for the oxidized periphery region for the oblique incidence of light at the angle 10 degrees, the angle being defined for a Ga.sub.0.85Al.sub.0.15As layer. In FIGS. 5A through 5C a double stage shift of the dips is shown. First, the two dips of the optical power reflectance spectrum of the VCSEL shift towards shorter wavelengths due to replacement of semiconductor layer or layers by oxide layer(s) having a lower refractive index. Dash-dotted tilted straight lines connect the corresponding features in FIGS. 5A and 5B. Second, the two dips further shift towards shorter wavelengths because of a tilt angle of incidence of light. Dash-dotted curves connect the corresponding features in FIGS. 5B and 5C. The angle of incidence is chosen such that the long wavelength dip in FIG. 5C is positioned at 850 nm. The vertical dashed line demonstrates, that the dip in the optical power reflectance spectrum of the VCSEL structure calculated for the aperture core region at normal incidence, on the one hand, and the long wavelength dip in the optical reflectance spectrum calculated for the oxidized periphery region at oblique incidence, on the other hand, coincide.

[0043] FIGS. 6A through 6D compare the electric field strength profiles for the optical modes of the vertical cavity surface emitting laser (VCSEL) of FIG. 4A, calculated for the same wavelength of 850 nm, one mode being the vertical optical mode in the non-oxidized core region, and the other being the tilted optical mode in the oxidized periphery region. FIG. 6A shows the refractive index profile in the non-oxidized core region.

[0044] FIG. 6B depicts the electric field strength profile of the vertical optical mode of the core region of the VCSEL of FIG. 4A.

[0045] FIG. 6C demonstrates the electric field strength profile of the optical mode of the VCSEL of FIG. 4A calculated for the periphery region and the mode tilt angle of 10 degrees, defined for a Ga.sub.0.85Al.sub.0.15As layer. Each of the nodes of the electric field profile of the tilted mode of FIG. 6C is connected by a dashed line to the nearest node of the electric field profile of the vertical mode of FIG. 6B. The vertical mode has one extra node. However, difference in refractive index profiles in the non-oxidized core region and in the oxidized periphery region results in redistribution of the electric field strength profile, and hence in the break of orthogonality between the two modes of FIGS. 6B and 6C, thus enabling leakage of the optical modes from the core to the periphery.

[0046] FIG. 6D yields the refractive index profile in the oxide region of the VCSEL of FIG. 4A.

[0047] FIG. 7A. Schematic cross-section of a vertical cavity surface emitting laser (VCSEL) with coupled cavity.

[0048] FIG. 7B shows schematically dispersion curves of the VCSEL mode in the core region of the VCSEL of FIG. 7A and one of the tilted modes, namely, of the second cavity mode of the periphery region of a planar multilayer structures, mimicking the core and the periphery regions of the VCSEL of FIG. 7A. VCSEL modes can leak from the core to the periphery.

[0049] FIG. 8A shows schematically a cross-section of a vertical cavity surface emitting laser (VCSEL) subject to a selective chemical transformation process resulting in a transformation of the chemical profile in the periphery region, according to an embodiment of the present invention.

[0050] FIG. 8B shows schematically dispersion curves of a VCSEL optical mode of the core region and of a tilted optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the VCSEL of FIG. 8A. The transformation of the chemical profile in the periphery region enables leakage of the optical modes from the core to the periphery.

[0051] FIGS. 9A through 9D illustrate schematically different ways of the chemical transformation of a semiconductor alloy subject to impurity diffusion. FIG. 9A shows schematically the alloy composition profile of a multilayer alloy structure in the structure region not subject to the chemical transformation (core region).

[0052] FIG. 9B refers to the case where the chemical transformation is an alloy compositional intermixing evolving towards formation of a homogeneous alloy. FIG. 9B illustrates an intermediate stage of the alloy compositional intermixing in which the alloy composition profile reduces its amplitude and smoothens as compared to FIG. 9A.

[0053] FIG. 9C shows schematically the alloy composition profile of a multilayer alloy structure in the structure region not subject to the chemical transformation (core region) repeating FIG. 9A.

[0054] FIG. 9D refers to the case where a particular chemical transformation is the phase separation of the alloy. FIG. 9D illustrates an intermediate stage of the phase separation wherein the amplitude of alloy composition variations have enhanced as compared to the initial state of FIG. 9C.

[0055] FIG. 10 shows schematically the I-V curve of the p-n junction containing the quantum well active medium (top panel) and the I-V curve of the p-n junction in the passive region at the periphery between the non-transformed and transformed parts of the periphery (bottom panel).

[0056] FIGS. 11A through 11D compare the electric field strength profiles for the optical modes of the vertical cavity surface emitting laser (VCSEL) of FIG. 8A, calculated for the same wavelength of 850 nm, one mode being the vertical optical mode in the core region, and the other being the vertical optical mode in the periphery region in which chemical transformation has occurred. FIG. 11A shows the refractive index profile in the core region (not intermixed).

[0057] FIG. 11B depicts the electric field strength profile of the vertical optical mode of the core region of the VCSEL of FIG. 8A.

[0058] FIG. 11C demonstrates the electric field strength profile of the optical mode of the VCSEL of FIG. 8A calculated for the periphery region in which chemical transformation has occurred, and such chemical transformation is alloy compositional intermixing. The mode profile repeats the profile of the VCSEL optical mode of FIG. 11B in the non-intermixed layers of the periphery and is a combination of two travellin planar waves in the intermixed layers of the periphery. Each of the local field minima of the field profile of FIG. 11C is connected by a dashed line to the nearest local field minimum of the field profile of FIG. 11B. Both field profiles have the same number of nodes. However, as the profile of FIG. 11C in the intermixed layers of the periphery has a different shape than the profile of FIG. 11B in the same layers in the core, the overlap between the two profiles is less than 100%.

[0059] FIG. 11D yields the refractive index profile in the partially intermixed periphery region of the VCSEL of FIG. 8A.

[0060] FIG. 12A shows schematically a cross-section of a vertical cavity surface emitting laser (VCSEL) subject to a transformation of the chemical profile in the periphery region, wherein additional oxide-confined aperture is introduced to determine the current path.

[0061] FIG. 12B shows schematically dispersion curves of a VCSEL optical mode of the core region and of a tilted optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the VCSEL of FIG. 12A. The transformation of the chemical profile in the periphery region enables leakage of the optical modes from the core to the periphery.

[0062] FIG. 13A shows schematically a cross-section of a vertical cavity surface emitting laser (VCSEL) subject to transformation of the chemical profile in the periphery region, wherein the transformation occurs through the active medium down to the bottom distributed Bragg reflector, according to yet another embodiment of the present invention.

[0063] FIG. 13B shows schematically dispersion curves of a VCSEL optical mode of the core region and of a tilted optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the VCSEL of FIG. 13A. The transformation of the chemical profile in the periphery region enables leakage of the optical modes from the core to the periphery.

[0064] FIG. 14A shows schematically a cross-section of a vertical cavity surface emitting laser (VCSEL) subject to transformation of the chemical profile in the periphery region, wherein several layers of selected alloy composition are added in the region where a p-n junction between the non-transformed and transformed parts of the periphery is formed, aimed at an increase of the opening voltage, according to another embodiment of the present invention,

[0065] FIG. 14B shows schematically dispersion curves of a VCSEL optical mode of the core region and of a tilted optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the VCSEL of FIG. 14A. The transformation of the chemical profile in the periphery region enables leakage of the optical modes from the core to the periphery.

[0066] FIG. 15 shows the energy band diagram of the Ga.sub.1-xAl.sub.xAs alloys showing the maximum energy band gap at the aluminum composition close to x=0.45 corresponding to the cross-over point between direct Γ-minimum and indirect X-minimum of the conduction band.

[0067] FIG. 16 shows schematically a plan view of phase-coupled array of vertical cavity surface emitting lasers (VCSELs), wherein the coupling is realized via the leakage of the optical modes from the core regions into the surrounding periphery regions, according to another embodiment of the present invention.

[0068] FIG. 17A shows schematically a plan view of phase-coupled array of vertical cavity surface emitting lasers (VCSELs), wherein the coupling is realized via the leakage of the optical modes from the core regions into the surrounding periphery regions, similar to FIG. 16, and further denotes a plane for the vertical cross-section to be shown in FIG. 17B.

[0069] FIG. 17B shows schematically a vertical cross-section of a phase-coupled array of vertical cavity surface emitting lasers of FIGS. 16 and 17A, wherein such array is employed for the steering of the laser beam.

[0070] FIG. 18A shows schematically a cross-sectional view of a striped edge-emitting laser, in which areas adjacent to the stripe are subject to chemical transformation realized as alloy compositional intermixing, according to yet another embodiment of the present invention.

[0071] FIG. 18B shows schematically a top view of the edge-emitting laser of FIG. 18A.

[0072] FIGS. 19A through 19G illustrate the operational principle of the edge-emitting laser of FIGS. 18A and 18B. FIG. 19A shows the vertical refractive index profile of the edge-emitting laser of FIG. 18A in the core region where no chemical transformation has occurred. The profile contains two coupled waveguides, and the quantum well-based active medium is located in the bottom (active) waveguide.

[0073] FIG. 19B shows the vertical profile of the mode #1 (fundamental mode) of the structure of FIG. 19A. The mode is predominantly located in the passive waveguide.

[0074] FIG. 19C shows the vertical profile of the mode #2 (a next high-order mode) of the structure of FIG. 19A. The mode is predominantly located in the active waveguide. This is the mode having the maximum optical confinement factor in the active medium, i.e. this is the lasing mode.

[0075] FIG. 19D shows the dispersion curves of the modes #1 and #2 in the planar multilayer structure mimicking the core of the edge-emitting laser of FIG. 18A.

[0076] FIG. 19E shows the vertical refractive index profile of the edge-emitting laser of FIG. 18A in the periphery region, in which alloy compositional intermixing has occurred resulting in the vanishing of the passive waveguide and formation of a new top cladding having a significantly smaller refractive index step with respect to the active waveguide.

[0077] FIG. 19F shows the vertical profile of the mode #1 (fundamental mode) of the structure of FIG. 19E. The fundamental mode in the periphery is predominantly located in the active waveguide.

[0078] FIG. 19G shows the dispersion curve of the mode #1 in the planar multilayer structure mimicking the periphery of the edge-emitting laser of FIG. 18A. The dispersion curve is located at larger angles compared with the dispersion curve of the lasing mode #2 in the core. According to FIGS. 4B, 7B, 8B, 12B, and 14B this implies that the lasing mode #2 can leak from the core to the periphery.

[0079] FIG. 20 shows schematically a plan view of a phase-coupled array of edge-emitting lasers, wherein the coupling is realized via the leakage of the optical modes from the core regions in the surrounding periphery regions, according to a further embodiment of the present invention.

[0080] FIG. 21A shows schematically a cross-section of a duo-cavity vertical cavity surface emitting laser (VCSEL) subject to transformation of the chemical profile in the periphery region, wherein the transformation is the alloy compositional intermixing and it occurs through the active medium down to the bottom distributed Bragg reflector, according to yet another embodiment of the present invention.

[0081] FIG. 21B shows schematically dispersion curves of a VCSEL optical mode of the core region and of the VCSEL optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the VCSEL of FIG. 21A. The alloy compositional intermixing in the periphery region results in a shift of the VCSEL optical mode towards longer wavelengths and enables leakage of the optical modes from the core to the periphery.

[0082] FIG. 22A shows the optical power reflectance spectrum at normal incidence of the vertical cavity surface emitting laser (VCSEL) structure of FIG. 21A in the core region revealing a VCSEL dip at 850 nm.

[0083] FIG. 22B shows the optical power reflectance spectrum at normal incidence of the VCSEL structure of FIG. 21A calculated for the periphery region upon alloy compositional intermixing. Vanishing of the second cavity due to compositional intermixing waves the resonant interaction between the active cavity and the second cavity and results in a long wavelength shift of the reflectivity dip of the active cavity.

[0084] FIG. 22C shows the optical power reflectance spectrum of the VCSEL structure of FIG. 21A calculated for the periphery region for the oblique incidence of light at the angle 9.3 degrees, the angle being defined for a Ga.sub.0.85Al.sub.0.15As layer. In FIGS. 22A through 22C a double stage shift of the dips is shown. First, the VCSEL dip of the active cavity shifts towards longer wavelengths due to vanishing of the second cavity upon alloy compositional intermixing. Dash-dotted tilted straight line connects the corresponding features in FIGS. 22A and 22B. Second, the dip shifts towards shorter wavelengths because of a tilt angle of incidence of light. Dash-dotted curves connect the corresponding features in FIGS. 22B and 22C. The angle of incidence is chosen such that the overall shift is zero and the dip calculated for the oblique incidence in the periphery region matches the dip calculated for the core region at normal incidence (at 850 nm). The vertical dashed line demonstrates that the dip in the optical power reflectance spectrum of the VCSEL structure calculated for the core region at normal incidence, on the one hand, and the dip in the optical reflectance spectrum calculated for the periphery region at oblique incidence, on the other hand, coincide.

[0085] FIGS. 23A through 23D compare the electric field strength profiles for the optical modes of the vertical cavity surface emitting laser (VCSEL) of FIG. 23A, calculated for the same wavelength of 850 nm, one mode being the vertical optical mode in the core region, and the other being the vertical optical mode in the periphery region. FIG. 23A shows the refractive index profile in the core region.

[0086] FIG. 23B depicts the electric field strength profile of the vertical optical mode of the core region of the VCSEL of FIG. 21A.

[0087] FIG. 23C demonstrates the electric field strength profile of the optical mode of the VCSEL of FIG. 21A calculated for the periphery region and the mode tilt angle of 9.3 degrees, defined for a Ga.sub.0.85Al.sub.0.15As layer. Each of the nodes of the electric field profile of the vertical mode of FIG. 23C is connected by a dashed line to the nearest node of the electric field profile of the vertical mode of FIG. 23B. Both profiles have the same number of nodes, and, hence, a large overlap integral.

[0088] FIG. 23D yields the refractive index profile in the oxide region of the VCSEL of FIG. 21A.

[0089] FIG. 24A shows schematically a cross-section of a passive cavity surface-emitting laser subject to transformation of the chemical profile in the periphery region, wherein the active medium is places in the top distributed Bragg reflector (DBR), and the resonant cavity is passive, and where the transformation is the alloy compositional intermixing and it occurs through the active medium down to the part of the top Bragg reflector, whereas the passive cavity is not subject to intermixing.

[0090] FIG. 24B shows schematically dispersion curves of the vertical optical mode in the core region and of a tilted optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the passive cavity surface emitting laser of FIG. 24A. The transformation of the chemical profile in the periphery region enables leakage of the optical modes from the core to the periphery.

[0091] FIG. 25A shows schematically a cross-section of a duo cavity passive cavity surface-emitting laser, emitting laser subject to transformation of the chemical profile in the periphery region, wherein the transformation is the alloy compositional intermixing and it occurs through the second cavity and through the active medium down to a part the bottom distributed Bragg reflector, wherein no transformation occurs in the first passive cavity, according to yet another embodiment of the present invention.

[0092] FIG. 25B shows schematically dispersion curves of a vertical optical mode of the core region and of the vertical optical mode of the periphery region of a planar multilayer structure mimicking the core and the periphery regions of the passive cavity surface emitting laser of FIG. 25A. The alloy compositional intermixing in the periphery region results in a shift of the vertical optical mode towards longer wavelengths and enables leakage of the optical modes from the core to the periphery.

DETAILED DESCRIPTION OF THE INVENTION

[0093] The present patent application discloses an alternative way to enhance leakage losses of the high-order transverse optical modes and thus to promote single transverse mode lasing. Technology of selective chemical transformation of the semiconductor structure is known in the Art. There are several processes that can be activated beneath the openings in a lithographic mask. One process is implantation of impurities. Another process includes diffusion of impurities from the surface. Such elements like Zinc or Silicon are commonly used. Yet another process includes implantation of point defects (an example is implantation of excess ions of Gallium). A further process includes diffusion of point defects like vacancies from the surface. Another process includes selective laser annealing. The purpose of all these properties is to selectively alter the chemical profile in the vertical directions beneath selective areas on the surface such that a domain of chemical transformation occurs. The side boundaries of such domain are controlled by the mask, the top boundary if commonly the surface of the structure (but also can be lower), and the bottom boundary can be varied by choosing particular regimes like temperature of the process, amount of excess atoms introduced, duration etc.

[0094] One of the commonly known type of chemical transformation of this type is alloy compositional intermixing, wherein a structure starting from alloy composition profile in the vertical direction evolves towards an alloy with the uniform composition. Alloy compositional intermixing is frequently applied once there is, for example, a need to form a periphery region of a semiconductor material having a broad band gap surrounding a core structure with a narrower band gap.

[0095] In the present disclosure a new application of the process of selective chemical transformation is employed. Selective chemical transformation is applied to control the optical modes of the laser structure. FIG. 8A shows schematically a cross-section of a VCSEL (800) according to an embodiment of the present invention. Selective chemical transformation is applied to a part of the top DBR (806), resulting in the formation of a chemically transformed region (890). The chemically transformed region determines the core of the device (850) and the periphery (860) in which the chemical transformation occurs. The transformed region (890) is thus the top part of the device periphery. On the bottom the transformed region (890) is bounded by the boundary (895) separating the transformed region (890) and non-transformed part of the top DBR (806). The side boundary (896) separates the core (850) of the device and the transformed region (890).

[0096] In the embodiment (800), the bottom DBR (102) is n-doped, and the top DBR (806) is p-doped. The transformed region (890) is formed by diffusion of a donor impurity, or an n-impurity. A one skilled in the art will appreciate that silicon can be applied as silicon introduced into a semiconductor structure formed of III-V semiconductor materials is an n-impurity. As initially a part of the top DBR (806) was p-doped, it is important that the concentration of the n-impurity exceeds the initial level of p-doping overcompensating the material rendering it n-doped. The dielectric layer (880) separates the n-doped transformed region (890) from the p-contact (112).

[0097] Once in a part of the periphery (860), namely in the region (890), chemical transformation has occurred, the refractive index profile in the vertical direction at the periphery (860) is different from the refractive index profile in the vertical direction in the core (850). Therefore, similarly to co-pending patent application U.S. Ser. No. 13/771,875, the orthogonality between the VCSEL mode in the core and a one or several tilted modes at the periphery is broken, which enables leakage of the VCSEL mode from the core to the periphery as illustrated in FIG. 8B.

[0098] FIGS. 9A and 9B illustrate a possible change of the alloy composition profile upon chemical transformation for the case, where such chemical transformation is alloy compositional intermixing. Once the structure, prior to intermixing is composed of an alloy material A.sub.1-cB.sub.cD (Ga.sub.1-cAl.sub.cAs as an example) with alternating alloy composition, wherein layers having a composition c.sub.1 and a thickness d.sub.1 alternates with layers having a composition c.sub.2 and a thickness d.sub.2. Then, the compete alloy compositional intermixing results in the formation of a homogeneous alloy with the alloy composition

[00004] c _ = c 1 .Math. .Math. d 1 + c 2 .Math. d 2 d 1 + d 2 . ( 9 )

Once the alloy compositional intermixing is not complete, alloy composition remains modulated, but its profile smoothens and the modulation amplitude decreases as shown in FIG. 9B.

[0099] A one skilled in the art will appreciate that some semiconductor alloys have a miscibility gap, wherein a homogeneous alloy in the certain range of compositions and at certain temperatures is thermodynamically unstable. Alloys Al.sub.1-xIn.sub.xAs and Ga.sub.1-xIn.sub.xAs.sub.1-yP.sub.y are known as examples. In this case processes like implantation or diffusion of impurities which enhances the bulk mobility of the main atoms constituent the semiconductor materials activate the trend to alloy phase separation. FIGS. 9C and 9D illustrate this case. The alloy composition profile in FIG. 9C refers to the structure prior to diffusion of impurities. Upon diffusion of impurities, phase separation process starts, and the amplitude of composition modulation increases as illustrated in FIG. 9D.

[0100] Alloy phase separation, also as alloy compositional intermixing results in a change of the refractive index profile and in breakdown of the orthogonality of the optical modes in two neighboring parts of the structure (core and periphery) and thus enabling leakage of the optical modes from the core to the periphery and thus promoting single transverse mode lasing.

[0101] FIG. 10 explains how the current flow occurs through the VCSEL (800) of FIG. 8A. The p-n junction located between the p-doped top DBR (806) and the n-doped bottom DBR (102) contains the active medium. The I-V curve of this p-n junction is shown on the top panel of FIG. 10. Upon applying a forward bias (113), the I-V curve opens once the voltage exceeds the opening voltage V.sub.0. A one skilled in the art will agree that the opening voltage is determined by the energy spectrum of the active medium and, approximately, by the photon energy of the emitted laser light,


V.sub.0=E.sub.photon/|e|,   (10)

where |e| is the elementary charge. The I-V curve of the second p-n junction existing between the n-doped chemically transformed region (890) and the p-doped DBR (806) is not biased, and there is no current flow through the boundaries (895) and (896) between the chemically transformed region (890) and the top DBR (806). Thus, the current flows only in the core region (850), and no generation of the light occurs beneath the top contact.

[0102] FIGS. 11A through 11D compare the two optical modes of the VCSEL structure of FIG. 8A calculated for the same wavelength 850 nm. FIG. 11A shows the vertical refractive index profile in the core region of the device. FIG. 11B shows the vertical electric field strength profile of the vertical optical mode of the core of the device. FIG. 11D shows the vertical refractive index profile in the periphery region of the device. The profile of FIG. 11D includes the region in which the alloy compositional intermixing has occurred forming a homogeneous alloy with a constant refractive index having some intermediate value between the low index and the high index of the DBR. Once the average alloy composition is conserved upon intermixing, the refractive index of the intermixed layer approximately equals to the averaged refractive index of the DBR. Thus the spectral position of the vertical mode remains nearly unchanged upon intermixing. FIG. 11C depicts the vertical electric strength profile of the vertical mode in the periphery region which has undergone alloy compositional intermixing.

[0103] It should be noted that the plot of FIG. 11C is inverted. The absolute values of the optical field strength that are the positive numbers are displayed in the downward vertical direction. This is done for convenience. In such a way it is more convenient to compare the nodes of the two optical field profiles. To make such a comparison, each node (or each local minimum) of the optical field profile of FIG. 11C is connected by a dashed line with the closest node of the optical field profile of FIG. 11B. One can notice that the optical field profiles of FIGS. 11B and 11C contain the same number of the nodes. However, the profile of FIG. 11C is drastically changed compared to that of FIG. 11B. The optical field profile of FIG. 11C in the non-transformed part of the periphery, namely in the bottom DBR, in the resonant cavity and in the small part of the top DBR adjacent to the cavity repeats that of FIG. 11b, but is reduced by a certain factor, whereas the field profile in the transformed (intermixed) part of the periphery is a combination of the outwards travelling plane wave and the travelling wave reflected from the semiconductor/air interface. It should be noted that the optical field profiles of FIG. 11B and FIG. 11C have equal number of nodes. However, due to the transformation of the field profile in the intermixed region, the overlap integral of these two modes is less than 100%. Evaluation of the overlap integral for the particular profiles of FIGS. 11B and 11C yields the overlap of 80% (two definitions of Eqs. (7) and (8) give even closer values than those for the fields of FIGS. 6B and 6C, with a difference below 1%, namely W.sub.ji.sup.(1)=80.1%, and W.sub.ji.sup.(2)=79.5%).

[0104] A one skilled in the art will appreciate that, for a given optical mode of the core, the sum of the overlap integrals of this mode with all the modes of the periphery always remains 100%. Therefore, once the overlap integral of the vertical mode in the core (FIG. 11B) with the vertical mode of the periphery (FIG. 11C) reduces from 100% down to 80% the sum of the overlap integrals of the vertical mode of the core with all the rest modes, i.e. with all the tilted modes of the periphery increases up to 20%. Depending on particular epitaxial structure and particular depth of the alloy compositional intermixing there may exist cases where the overlap integral of the vertical mode of the core with a single tilted mode of the periphery is large, like in FIGS. 6B and 6C. In other cases, the overlap integral of the vertical mode of the core with every single tilted mode of the periphery is small, but because of a large number of tilted modes, the overall leakage losses of the vertical mode of the core are significant.

[0105] FIG. 12 shows a schematic cross-section of a VCSEL (1200) according to another embodiment of the present invention. The chemical transformation is activated by the diffusion of a p-type impurity into the p-doped top DBR. Thus the chemically transformed region (1290) is doped by a p-type impurity. In the embodiment of FIG. 12 the p-type impurity is Zinc. In order to form the targeted path for the current, oxide confined aperture (1245) is formed by the selective oxidation of Ga.sub.1-xAl.sub.xAs layer, with a high composition of Aluminum (preferably 95% or higher) and its transformation into an amorphous oxide of Ga.sub.1-xAl.sub.xO.sub.y layer. As the oxide layer is an electric insulator, no current flow occurs through the active medium beneath the top contacts. As the boundaries (1295) and (1296) of the transformed region (1290) are the boundaries between the p-doped DBR (1206) and the p-doped transformed region (1290), no p-n junction is formed at the boundary.

[0106] A one skilled in the art will appreciate that the lateral structures formed by the chemical transformation (1290) and by selective oxidation of (1245) may form two different lateral dimensions, and a certain relation between two diameters should be met to promote proper functionality of the device. If the diameter of the oxide confined aperture (1245) is significantly smaller (by 1 micrometer or more) than the diameter of the core structure (1250), the optical modes are controlled by the oxide aperture but not by the chemically transformed region (1290). On the other hand, the oxide aperture should not be too large, to keep the capacitance of the device low. Thus, the preferred diameter of the oxide confined aperture (1245) is the same as the diameter of the non-transformed core (1250) or slightly larger (up to 1 micrometer larger).

[0107] A change of the vertical refractive index profile at the periphery (1260) with respect to the vertical refractive index profile in the core (1250) enables leakage of the optical modes from the core (1250) to the periphery (1260) and thus promotes a single transverse mode lasing (1235) (FIG. 12B).

[0108] FIG. 13A shows schematically a cross-section of a VCSEL (1300) according to yet another embodiment of the present invention. The chemically transformed region (1390) expands from the top surface of the device down through the top DBR (806), through the cavity (103) and ends within the bottom DBR (102). Once the top DBR (806) is p-doped, the bottom DBR (102) is n-doped, the intermixed region (1390) should be p-doped. This can be realized if the chemical transformation is activated by diffusion of a p-type impurity, e. g. Zinc. The boundaries (1395) and (1396) are the boundaries between the p-doped transformed region (1390) and n-doped bottom DBR (102) which implies a p-n junction at those boundaries. Thus the device (1300) contains two p-n junctions. A first p-n junction is formed between the n-doped bottom DBR (102) and the p-doped top DBR (806), wherein the active medium (105) is just located within the first p-n junction. And the second p-n junction is formed at the boundaries (1395) and (1396) between the n-doped bottom DBR (102) and the p-doped transformed region (1390).

[0109] FIG. 10 illustrates the operation of the device having two p-n junctions. The top panel of FIG. 10 shows the I-V curve of the first p-n junction wherein the junction opens at an opening voltage V.sub.0. The value of the opening voltage is determined by the energy of photons emitted by the device as mentioned above, in Eq. (10). On the other hand, the opening voltage of the second p-n junction equals


V.sub.1=E.sub.g/|e|,   (11)

where E.sub.g is the energy band gap of the semiconductor material at the boundaries (1395) and (1396). As the passive semiconductor materials are selected such that they are transparent to the emitted light, E.sub.g>E.sub.photon, and V.sub.1>V.sub.0. Thus, there exists an interval of bias voltages at which the first p-n junction is open, and the second p-n junction is still closed. Then the current flow occurs only in the core region (1350), and no current flows through the periphery region (1360). Hence, no light is generated beneath the top contact.

[0110] As the chemical transformation in the region (1390) results in a change of the vertical profile of the refractive index in the periphery region (1360) which is distinct from the vertical profile of the refractive index in the core (1350), the optical modes in the core and at the periphery are no longer orthogonal, which enables leakage of the optical modes from the core (1350) to the periphery (1360) as illustrated in FIG. 13B. This promotes single transverse mode lasing (1335).

[0111] To enable correct operation of the device in a broad interval of the bias voltages, one needs to render the opening voltage of the second p-n junction V.sub.1 as high as possible. FIG. 14A shows schematically a cross section of a VCSEL (1400) according to a further embodiment of the present invention. A p-doped chemically transformed region (1490) is formed extending from the top surface of the device through the top p-doped DBR (806), through the cavity (103), down to the n-doped bottom DBR (102). Boundaries (1495) and (1496) of the intermixed region (1490) form a second p-n junction. A special barrier region (1485) is introduced within the bottom DBR (102) such that the boundary (1495) between the intermixed region (1490) and the bottom DBR (102) lies within the region (1485) and that the layers of the barrier region (1485) has an energy bandgap broader than that of the rest of the layers in the structure. It is preferred that at least one layer of the region (1485) has the energy band gap exceeding the energy band gap of the rest of the semiconductor materials by at least 0.1 eV.

[0112] As the chemical transformation in the region (1490) results in a change of the vertical profile of the refractive index in the periphery region (1460) which is distinct from the vertical profile of the refractive index in the core (1450), the optical modes in the core and at the periphery are no longer orthogonal, which enables leakage of the optical modes from the core (1450) to the periphery (1460) as illustrated in FIG. 13B. This promotes single transverse mode lasing (1435).

[0113] FIG. 15 shows an energy band diagram of the semiconductor alloys Ga.sub.1-xAl.sub.xAs. A one skilled in the art will appreciate that, once a DBR is formed of alternating layers of Ga.sub.1-xAl.sub.xAs, typically an alloy Ga.sub.1-xAl.sub.xAs with the Aluminum molar fraction below 20% is chosen as high index material of the DBR. This can be, for example, Ga.sub.1-xAl.sub.xAs with Aluminum composition 15%. If the device is configured to emit light at a wavelength longer than 870 nm, at which GaAs is transparent, then GaAs can be used as a high index material of the DBR. As the band gap energy of the Ga.sub.1-xAl.sub.xAs alloy increases rapidly once the Aluminum composition increases from 0 to the cross-over point from the direct to indirect gap at x=45% and changes only slowly afterwards, the most important is to properly choose a high index material for the barrier region (1485) in FIG. 14A. To keep the functionality of the barrier region (1485) as a part of the DBR, i.e to keep certain refractive index step, on the one hand, and to increase the energy band gap of the material, on the other hand, it is preferred to choose, as a high index material, an alloy Ga.sub.1-xAl.sub.xAs with Aluminum molar fraction exceeding thirty (30) percent.

[0114] FIGS. 21A through 23D disclose another embodiment of the present invention. FIG. 21A shows schematically a cross-section of a duo cavity VCSEL (2100). A second cavity (2191) is placed within the top DBR (2106). The chemical transformation of the region (2190) is alloy compositional intermixing. The intermixed region (2190) is created such that the bottom boundary (2195) is placed above the active cavity (103) but below the second cavity (2191). The intermixed region (2190) is n-doped, and as no bias is applied to the p-n junction at the boundaries (2195) and (2196) of the intermixing region (2190), this junction is closed and no electric current flows through the intermixed region (2190). The n-doped intermixing region (2190) is covered by a dielectric layer (2180) which separates it from the p-contact (112).

[0115] FIGS. 22A through 22C show how the resonant interaction of two cavities manifests itself in the optical power reflectance spectra. Second cavity (2191) is configured such that its resonance wavelength is at a longer wavelength compared to the resonance of the active cavity. Additionally the resonance interaction between two cavities results in the repulsion of the two resonances from the spectral positions which they would have in the isolated cavities. Thus, FIG. 22A represents the optical power reflectance spectra of the VCSEL structure (2100) in the core region (2150) at normal incidence of light revealing two dips within the reflectivity stopband, one at 850 nm, and another one at 872 nm. In the periphery region the second cavity (2191) is compositionally intermixed with the neighboring DBR layers and disappears. Therefore, the repulsion of two resonances due to anticrossing vanishes, and the active cavity resonance shifts back to the spectral position of the isolated active cavity resonance, i.e. shifts towards a longer wavelength (863 nm in FIG. 22B) (a dash-dotted line connecting features in FIGS. 22A and 22B). Consequently, the optical power reflectance spectrum of the periphery calculated for an oblique incidence of light (at the angle 9.3° defined in the layer Ga.sub.0.85Al.sub.0.15As) reveals the dip matching the VCSEL dip at normal incidence in the core at 850 nm.

[0116] Correspondingly, FIG. 21B shows that the VCSEL mode of the core lies within the continuum spectrum associated with the vertical mode of the periphery, and that leakage of the VCSEL mode from the core (2150) to the periphery (2160) is allowed at all wavelengths. This leakage promotes single transverse mode lasing (2135) of the VCSEL (2100) of FIG. 21A.

[0117] FIGS. 23A through 23D compare the two optical modes of the VCSEL structure of FIG. 21A calculated for the same wavelength 850 nm. FIG. 23A shows the vertical refractive index profile in the core region (2150) of the device (2100) containing two coupled cavities. FIG. 23B shows the vertical electric field strength profile of the vertical optical mode of the core of the device. FIG. 23D shows the vertical refractive index profile in the periphery region (2160) of the device (2100) wherein the second cavity has vanished due to alloy compositional intermixing. The profile of FIG. 23D includes the intermixed region in which the complete alloy compositional intermixing has occurred forming a homogeneous alloy with a constant refractive index having some intermediate value between the low index and the high index of the DBR. FIG. 23C depicts the vertical electric strength profile of the vertical mode in the periphery region which has undergone intermixing.

[0118] It should be noted that the plot of FIG. 23C is inverted. The absolute values of the optical field strength that are the positive numbers are displayed in the downward vertical direction. This is done for convenience. In such a way it is more convenient to compare the nodes of the two optical field profiles. To make such a comparison, each node (or each local minimum) of the optical field profile of FIG. 23C is connected by a dashed line with the closest node of the optical field profile of FIG. 23B. One can notice that the optical field profiles of FIGS. 23B and 23C contain the same number of the nodes which results in a large overlap integral between the fields of FIGS. 11B and 11C (the overlap equals 85%, again two definitions of Eqs. (7) and (8) give very close values: W.sub.ji.sup.(1)=85.3%, and W.sub.ji.sup.(2)=84.8%).

[0119] FIG. 24A shows schematically a cross-section of a passive cavity surface emitting laser (2400) according to yet another embodiment of the present invention. The concept of the passive cavity laser including passive cavity surface emitting laser was disclosed in the U.S. Pat. No. 8,472,496 “OPTOELECTRONIC DEVICE AND METHOD OF MAKING SAME”, filed Jul. 6, 2010, issued Jun. 25, 2013, invented by one of the inventors of the present invention (Ledentsov), wherein the patent is incorporated herein by reference. In a passive cavity surface emitting laser the resonant cavity is passive, and the active medium is placed within one of the DBRs. Fabrication and characterization of a passive cavity surface-emitting laser was carried out in the publication (J. A. Lott, V. A. Shchukin, N. N. Ledentsov, A. M. Casten, and K. D. Choquette, “Passive cavity surface emitting laser”, Electronic Letters, Volume 41, Issue 12, pages 717-718, 9 Jun. 2011, wherein this publication is hereby incorporated herein by reference.) Despite a reduction of the optical confinement factor of the lasing optical mode of the device with respect to the value in a conventional VCSEL by approximately a factor of 3, the passive cavity surface emitting laser has demonstrated stable lasing.

[0120] The passive cavity surface emitting laser (2400) comprises an n-doped bottom DBR (102), an n-doped passive cavity (2403), and a top DBR (2406), wherein the latter consist of an n-doped section part (2465), an undoped section further comprising an active medium (2405), and a p-doped section (2466). The region (2490) in which chemical transformation occurs expands from the top surface down through the p-doped section (2466) of the top DBR (2465), through the active medium (2405) and has a bottom boundary (2495) within the n-doped section (2465) of the top DBR (2406). The chemical transformation in the region (2490) is activated by diffusion of a p-type impurity, preferably of Zinc.

[0121] Similar to the device (1300), the passive cavity surface emitting laser (2400) has two p-n junctions. A first p-n junction is formed between the n-doped part (2465) of the top DBR (2406) and a p-doped part (2466) of the top DBR (2406), wherein the active medium (2405) is located within the first p-n junction. The second p-n junction is formed at the boundaries (2495) and (2496) between the p-doped chemically transformed region (2490) and the n-doped section (2465) of the top DBR (2406). It was shown above in the description of the embodiment of FIG. 13 that in a certain interval of the bias voltage, the first p-n junction is open whereas the second p-n junction is closed, and no current flow occurs through the chemically transformed region (2490), and no light is generated beneath the top contact.

[0122] As the chemical transformation results in a change of the vertical profile of the refractive index in the periphery region (2460) which is distinct from the vertical profile of the refractive index in the core (2450), the optical modes in the core and at the periphery are no longer orthogonal, which enables leakage of the optical modes from the core (2450) to the periphery (2460) as illustrated in FIG. 24B and thus promotes single mode lasing (2435).

[0123] FIG. 25A shows schematically a duo cavity passive cavity surface emitting laser (2500) according to a further embodiment of the present invention. The device (2500) combines the properties of the two embodiments, the one of FIG. 21A and the one of FIG. 24A. The selective chemical transformation carried out in the region (2590) is controlled in such a way that the process is an alloy compositional intermixing activated by the diffusion of a p-type impurity, preferably of Zinc. The device (2500) as compared to the passive cavity surface-emitting laser (2400) contains, in the p-doped part (2566) of the top DBR (2506) also a second cavity (2591). The second cavity (2591) is present only in the core part (2550) of the device, whereas it has disappeared at the periphery (2560) due to alloy compositional intermixing. The duo cavity approach, as described for the embodiment of FIGS. 21A-23D, enables a strong leakage of the vertical modes of the core to the vertical modes of the periphery (as also shown in FIG. 25B), whereas the overlap integral of both modes is large. A further advantage of the passive cavity surface emitting laser (2500) versus VCSEL (2100) is a possibility to apply a p-type impurity, e.g. Zinc to activate alloy compositional intermixing.

[0124] In another group of embodiments of the present invention the structure of the device in the core region is configured such that not a vertical mode, but a tilted mode propagating at a certain non-zero angle with respect to the direction normal to the layers of the structure is the designated mode that should lase. Also for such a device, which is known as tilted cavity laser, the present invention can be applied to enhance the leakage losses of the undesired lateral modes, the leakage occurring to the periphery region of the structure. A one skilled in the art will appreciate that such a tilted cavity laser contains multilayer interference reflectors that are configured to promote reflection and feedback for the tilted optical mode propagating at a certain angle or in certain interval of angles. A distributed Bragg reflector is thus a particular realization of a multilayer interference reflector configured for the normal propagation of light.

[0125] FIG. 16 shows schematically a plan view of an array (1600) of phase-coupled VCSELs, according to yet another embodiment of the present invention. An array of holes (1450) is formed on the surface. The selective chemical transformation process is carried out. As a result of the process chemically transformed regions (1460) are formed around the holes. The width of the chemically transformed regions is preferably selected such that the chemically transformed regions formed around neighboring holes overlap or are close to overlap. The areas (1610) between the intermixed regions are the areas, in which the current flow through the active medium is possible and in which no chemical transformation occurs. The contact is deposited onto these areas (1610). The epitaxial structure is configured according to the present invention (FIGS. 8A, 12A, 13A, 14A, 21A, 24A, 25A). The optical modes excited beneath the contacts leak into the chemically transformed regions. Due to a sufficiently strong optical coupling between neighboring chemically transformed regions, a single coherent mode can be formed extending over the entire array of VCSELs. In a further embodiment of the present invention, a single coherent mode extends across the entire wafer.

[0126] FIGS. 17A and 17B shows schematically an array of phase coupled VCSELs, which is used for beam steering. FIG. 17A repeats FIG. 16, and contains additionally a line (1750) which defines a vertical plane.

[0127] FIG. 17B shows schematically a cross-section of the array of phase-coupled VCSELs (1600) in the vertical plane defined by the line (1750). The top contact (1612) is deposited on the top surface of the array, except the holes (1450). The contact area (1610) is depicted in FIG. 17B as three parts that are not electrically connected. Through each part, an independent current (J.sub.1, J.sub.2, or J.sub.3) flows. The paths of the current through neighboring sections (1620) is separated by the chemically transformed regions (1690). The current spreads and reaches the active cavity (1603), in which the active medium (1605) is placed. The bottom contact (1611) is preferably formed as a single contact for the whole array. The array of VCSELs (1600) is capable to emit a phase coupled laser light. The individually controllable currents (J.sub.1, J.sub.2, or J.sub.3) control the phase of the phase-coupled optical field of the laser light, and, hence, the direction, the vertical or tilted at a controlled angle, of the emitted coherent laser beam.

[0128] A one skilled in the art will appreciate, that, similar to an array of circular holes (1450), and array of elongated holes can be formed. This array can be configured such that it fixes polarization of a polarized laser light emitted by the array.

[0129] A one skilled in the art will further appreciate that an array similar to that of FIGS. 16 and 17 can be configured such that each pumped area of the array does not generate laser light but operates as a light-emitting diode or a gain chip. Then, once put into an external cavity, the array will generate laser light, similar to a Vertical-External-Cavity Surface-Emitting Laser (VECSEL).

[0130] Embodiments of the FIGS. 18A through 20 describe the same approach disclosed in the present application for a duo cavity VCSEL (FIG. 21A) or duo cavity passive cavity surface emitting laser (FIG. 25A) but employed for edge-emitting lasers.

[0131] FIG. 18A shows schematically a cross-section of an edge-emitting laser (1800) according to an embodiment of the present invention. The laser contains two coupled cavities, or, the same, two coupled waveguides, namely the active waveguide (1803) and the passive waveguide (1853), separated by a top cladding layer (1852) which is an evanescent reflector. The active cavity, or the active waveguide (1803) contains the active medium (1805). The active waveguide (1803) is bounded from the bottom by a bottom cladding (1802) which is an evanescent reflector. The structure is grown epitaxially on a substrate (1801). Selective diffusion of impurities followed that activates alloy compositional intermixing results in the formation of the intermixed regions (1890). In the embodiment of FIG. 18A, an n-type impurity, e.g. Silicon is used for the alloy compositional intermixing process. An n-contact (1811) is mounted on substrate (1801). A dielectric layer (1880) is deposited on top of the intermixed region (1890) to isolate electrically the n-doped region from the p-contact (1812).

[0132] FIG. 18B shows schematically a planar cross-section of the edge-emitting laser of FIG. 18A showing the core region (1850) and the periphery region (1860).

[0133] FIGS. 19A through 19G illustrate the operation of the edge-emitting laser (1800) configured, for definiteness, to emit laser light at the wavelength 980 nm. FIG. 19A displays the vertical refractive index profile of the laser (1800). The profile contains two coupled waveguides, namely the active waveguide further containing the active medium, and the passive waveguide. The both waveguides are configured such that the fundamental vertical mode denoted in FIG. 19B as Mode #1 is localized mainly in the passive waveguide, and the next order vertical mode denoted Mode #2 is localized mainly in the active waveguide (FIG. 19C). Mode #2 having the maximum optical confinement factor in the active medium is the mode in which laser light is generated. FIG. 19D shows dispersion curves of the Mode #1 and Mode #2. The dispersion curves are plotted in the same variables “angle”-“wavelength” and the dispersion curves (FIG. 3B) of the planar multilayer structure mimicking VCSEL (FIG. 3A). A one skilled in the art will appreciate that, in variables “angle”-“wavelength”, once the angle is fixed, the curve corresponding to a mode with a smaller mode number is positioned at a longer wavelength. Further, once the wavelength is fixed, the dispersion curve for the mode having a lower mode number is positioned at larger angles. Both the dispersion curves for a VCSEL-type structure (FIG. 3B) and for an edge-emitting structure obey thus rule.

[0134] FIG. 19E shows the vertical profile of the refractive index upon alloy compositional intermixing. It is assumed that the intermixing has occurred in the top cladding layer and in the passive waveguide. Once the passive waveguide has disappeared, the mode that was localized in the passive waveguide has vanished. The fundamental vertical mode at the periphery of the structure is now localized in the active waveguide (Mode #1P in FIG. 19F).

[0135] FIG. 19G shows the dispersion curve of Mode #1P in the periphery region of the structure.

[0136] A one skilled in the art will appreciate that, in the core of the structure, the resonant interaction of Mode #1 and Mode #2 results in a repulsion, whereas, for a given wavelength, Mode #1 shifts towards larger angles and Mode #2 shifts towards smaller angles. Once, at the periphery, one of the two modes vanishes, the remaining Mode #1P at the periphery, for a given wavelength, has an effective angle, intermediate between the values of the angles for the modes #1 and #2 in the core. Therefore, with respect to the lasing Mode #2 of the core, Mode #1′ of the periphery has a dispersion curve positioned at larger angles. Thus, the lasing mode is allowed to leak in the transverse lateral direction to the periphery, into the Mode #1P. This leakage promotes single transverse mode lasing of an edge-emitted laser (1800).

[0137] In a further embodiment of the present invention the same approach is applied to a tilted wave laser disclosed in the U.S. Pat. No. 7,421,001 “EXTERNAL CAVITY OPTOELECTRONIC DEVICE”, filed Jun. 16, 2006, issued Sep. 2, 2008, and U.S. Pat. No. 7,583,712 “OPTOELECTRONIC DEVICE AND METHOD OF MAKING SAME”, filed Jan. 3, 2007, issued Sep. 1, 2009, both by two of the three inventors of the present invention (Ledentsov and Shchukin), wherein these patents are hereby incorporated herein by reference. The approach disclosed in the present patent application enables fabrication of single transverse mode titled wave lasers.

[0138] In another embodiment of the present invention the approach disclosed in the present patent application is applied to a semiconductor disc laser.

[0139] In yet another embodiment of the present invention the approach disclosed in the present patent application can be applied to a single photon emitter.

[0140] In a further embodiment of the present invention the approach disclosed in the present patent application is applied to an emitter of entangled photons.

[0141] In another embodiment of the present invention the approach disclosed in the present patent application is applied to a light-emitting diode.

[0142] FIG. 20 shows schematically a plan view of a coherently coupled array (2000) of edge-emitters, according to another embodiment of the present invention. An array of holes (2080) is formed on the surface. The alloy compositional intermixing process is carried out. As a result of the process, the regions (2020) in which alloy compositional intermixing has occurred are formed around the holes. The width of the intermixed areas is preferably selected such that the intermixed regions formed around neighboring holes overlap forming extended intermixed areas shown approximately as grey rectangles. Current is injected through the contacts (2012). The epitaxial structure is designed according to the present invention (FIGS. 18A through 19G). The optical modes excited beneath the contacts leak into the intermixed regions. Due to a sufficiently strong coupling between neighboring intermixed regions, a single coherent mode can be formed across the array of edge-emitting lasers. In a further embodiment of the present invention, a single coherent mode extends across the entire wafer.

[0143] Although the consideration and modeling of example embodiments has been carried out for TE-optical modes, a one skilled in the art will appreciate that a similar consideration can be done for devices operating in TM-optical modes. Therefore, all disclosed features can be formulated in a general way for optical modes of any kind.

[0144] All publications, patents and patent applications mentioned in this specification are herein incorporated in their entirety by reference into the specification, to the same extent as if each individual publication, patent or patent application was specifically and individually indicated to be incorporated herein by reference. In addition, citation or identification of any reference in this application shall not be construed as an admission that such reference is available as prior art to the present invention.

[0145] Although the invention has been illustrated and described with respect to exemplary embodiments thereof, it should be understood by those skilled in the art that the foregoing and various other changes, omissions and additions may be made therein and thereto, without departing from the spirit and scope of the present invention. Therefore, the present invention should not be understood as limited to the specific embodiments set out above but to include all possible embodiments which can be embodied within a scope encompassed and equivalents thereof with respect to the feature set out in the appended claims.