Semiconductor Laser Module

Abstract

A semiconductor laser module comprises a tapered laser diode and/or a tapered amplifier diode equipped with beam shaping optics. The tapered laser diode and/or the tapered amplifier diode includes an emission facet for emitting a laser beam along a beam axis. The beam-shaping optics comprise a plano-convex cylindrical lens oriented so as to change divergence of the beam in the fast axis direction, the plano-convex spherical cylindrical lens having a planar surface arranged facing the facet and a circular cylindrical surface facing away from the facet. The refractive index of lens may be uniform throughout the entire lens. Alternatively, the lens may have a refractive index varying in the direction of the slow axis and/or in the direction of the fast axis.

Claims

1. A semiconductor laser module, comprising at least one of a laser diode and an amplifier diode, including an emission facet for emitting a laser beam along a beam axis, the laser beam having a fast axis direction and a slow axis direction; and beam shaping optics; wherein the beam-shaping optics comprise a plano-convex cylindrical lens having a planar surface arranged facing the facet, a circular cylindrical surface facing away from the facet, a cylinder axis oriented substantially parallel to the slow axis direction and a refractive index varying, at least, in the direction of the cylinder axis, the plano-convex cylindrical lens being configured to change divergence of the laser beam both in fast axis direction and in slow axis direction.

2. The semiconductor laser module as claimed in claim 1, wherein the laser beam has a beam quality factor in fast axis direction, hereinafter denoted M.sub.f.sup.2, and a beam quality factor in slow axis direction, hereinafter denoted M.sub.s.sup.2; and wherein the refractive index at the beam axis, hereinafter denoted n.sub.0, is chosen such that the condition $93 .Math. % .Math. \sqrt{\sin .Math. .Math. u_{f} / \sin (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})} \leq n \leq 107 .Math. % .Math. \sqrt{\sin .Math. .Math. u_{f} / \sin (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})},$ where u.sub.f denotes the uncorrected 1/e.sup.2 fast axis divergence half-angle of the laser beam and u.sub.s denotes the uncorrected 1/e.sup.2 slow axis divergence half-angle of the laser beam, is satisfied.

3. The semiconductor laser module as claimed in claim 2, wherein the plano-convex cylindrical lens is configured and positioned such that the circular cylindrical surface intersects the beam axis at an axial position where the ratio of the slow axis beam diameter of the laser beam to the fast axis beam diameter of the laser beam amounts to a value in the range from 0.87.Math.M.sub.s.sup.2/M.sub.f.sup.2 to 1.15.Math.M.sub.s.sup.2/M.sub.f.sup.2.

4. The semiconductor laser module as claimed in claim 2, wherein a radius of curvature of the circular cylindrical surface of the plano-convex cylindrical lens and a variation of the refractive index are chosen such that a ratio of the divergence of the beam in fast axis direction after the lens to the divergence of the beam in fast slow direction after the lens matches the ratio M.sub.f.sup.2/M.sub.s.sup.2.

5. The semiconductor laser module as claimed in claim 2, wherein the circular cylindrical surface of the plano-convex cylindrical lens has a radius of curvature satisfying the condition: $87 .Math. % .Math. \frac{d_{f}}{2 .Math. (n_{0} + 1) .Math. \tan (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})} \leq r \leq 115 .Math. % .Math. \frac{d_{f}}{2 .Math. (n_{0} + 1) .Math. \tan (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})}$ where r denotes the radius of curvature, d.sub.f denotes the beam diameter in fast axis direction of the laser beam at the axial position in which the circular cylindrical surface intersects the beam axis, n.sub.0 denotes the refractive index at the beam axis and u.sub.s denotes the uncorrected slow axis divergence half-angle of the laser beam.

6. The semiconductor laser module as claimed in claim 1, wherein the refractive index is highest at the beam axis and decreases in the direction of the cylinder axis with increasing distance from the beam axis, the variation of the refractive index in the direction of the cylinder axis being polynomial, or at least approximately polynomial, as a function of a slow axis coordinate.

7. The semiconductor laser module as claimed in claim 1, wherein the plano-convex cylindrical lens is configured to collimate the laser beam both in fast axis direction and in slow axis direction.

8. The semiconductor laser module as claimed in claim 1, wherein the plano-convex cylindrical lens a refractive index varying in the direction of the cylinder axis and in the fast axis direction.

9. The semiconductor laser module as claimed in claim 1, wherein the plano-convex cylindrical lens has an acceptance angle greater than or equal to the uncorrected 1/e.sup.2 slow axis divergence angle of the laser beam.

10. The semiconductor laser module as claimed in claim 1, wherein the laser beam has a first virtual source from which the laser beam appears to diverge in fast axis direction and a second virtual source from which the laser beam appears to diverge in slow axis direction, the second virtual source being located farther away from the emission facet than the first virtual source, wherein the plano-convex cylindrical lens changes the divergence of the beam in fast axis direction in such a way that a first virtual image of the first virtual source is produced and the divergence of the beam in slow axis direction in such a way that a second virtual image of the second virtual source is produced and wherein the first and second virtual imaged substantially coincide.

11. A semiconductor laser array comprising a stack of a plurality of semiconductor laser modules as claimed in claim 1.

12. A semiconductor laser module, comprising at least one of a tapered laser diode and a tapered amplifier diode, including an emission facet for emitting a laser beam along a beam axis; beam shaping optics; wherein the beam-shaping optics comprise a plano-convex cylindrical lens having a planar surface arranged facing the emission facet, a circular cylindrical surface facing away from the emission facet and a refractive index varying both in the slow axis direction and in the fast axis direction, the plano-convex cylindrical lens being configured to change divergence of the beam both in fast axis direction and in slow axis direction.

13. The semiconductor laser module as claimed in claim 12, wherein the laser beam has a beam quality factor in fast axis direction, hereinafter denoted M.sub.f.sup.2, and a beam quality factor in slow axis direction, hereinafter denoted M.sub.s.sup.2 and wherein the circular cylindrical surface has a radius of curvature and the plano-convex cylindrical lens a variation of the refractive index that are chosen such that a ratio of the divergence of the beam in fast axis direction after the lens to the divergence of the beam in fast slow direction after the lens matches the ratio M.sub.f.sup.2/M.sub.s.sup.2.

14. The semiconductor laser array as claimed in claim 12, wherein the plano-convex cylindrical lens is configured and positioned such that the circular cylindrical surface intersects the beam axis at an axial position where the ratio of the slow axis beam diameter of the laser beam to the fast axis beam diameter of the laser beam amounts to a value in the range from 0.87.Math.M.sub.s.sup.2/M.sub.f.sup.2 to 1.15.Math.M.sub.s.sup.2/M.sub.f.sup.2.

15. The semiconductor laser module as claimed in claim 12, wherein the plano-convex cylindrical lens has a refractive index at the beam axis, hereinafter denoted no, chosen such that the condition $93 .Math. % .Math. \sqrt{\sin .Math. .Math. u_{f} / \sin (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})} \leq n_{0} \leq 107 .Math. % .Math. \sqrt{\sin .Math. .Math. u_{f} / \sin (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})},$ where u.sub.f denotes the uncorrected 1/e.sup.2 fast axis divergence half-angle of the laser beam and u.sub.s denotes the uncorrected 1/e.sup.2 slow axis divergence half-angle of the laser beam, is satisfied.

16. The semiconductor laser module as claimed in claim 12, wherein the circular cylindrical surface of the plano-convex cylindrical lens has a radius of curvature r satisfying the condition: $87 .Math. % .Math. \frac{d_{f}}{2 .Math. (n + 1) .Math. \tan (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})} \leq r \leq 115 .Math. % .Math. \frac{d_{f}}{2 .Math. (n + 1) .Math. \tan (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})},$ where d.sub.f denotes the beam diameter in fast axis direction of the laser beam at the axial position in which the circular cylindrical surface intersects the beam axis, n.sub.0 denotes the refractive index at the beam axis and u.sub.s denotes the uncorrected slow axis divergence half-angle of the laser beam.

17. The semiconductor laser module as claimed in claim 12, wherein the refractive index is highest at the beam axis and decreases with increasing distance from the beam axis, the variation of the refractive index being polynomial, or at least approximately polynomial, as a function of a slow axis coordinate and a fast axis coordinate.

18. The semiconductor laser array as claimed in claim 12, wherein the plano-convex cylindrical lens is configured to collimate the laser beam both in fast axis direction and in slow axis direction.

19. The semiconductor laser module as claimed in claim 12, wherein the plano-convex cylindrical lens has an acceptance angle greater than or equal to the uncorrected 1/e.sup.2 slow axis divergence angle of the laser beam

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Preferred embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings, in which:

[0029] FIG. 1: is a perspective schematic view of a semiconductor laser module according to a first preferred embodiment of the invention;

[0030] FIG. 2: is a top schematic view of the semiconductor laser module of FIG. 1 with additional beam collimating optics;

[0031] FIG. 3: is a lateral schematic view of the system of FIG. 2;

[0032] FIG. 4: is an illustration of the aplanatic condition;

[0033] FIG. 5: is an illustration of beam refraction by a thick lens;

[0034] FIG. 6: is a perspective schematic view of a semiconductor laser diode array;

[0035] FIG. 7: is a perspective schematic view of a semiconductor laser module according to a further preferred embodiment of the invention;

[0036] FIG. 8: is a graph illustrating variation of the refractive-index in slow and fast axis direction of a gradient-index cylindrical lens used in an example embodiment of the invention;

[0037] FIG. 9: is a schematic view of the fast-axis plane of a semiconductor laser module featuring a gradient-index plano-convex spherical cylindrical lens;

[0038] FIG. 10: is a schematic view of the slow-axis plane of the semiconductor laser module of FIG. 9.

DESCRIPTION OF EMBODIMENTS

[0039] FIG. 1 illustrates a semiconductor laser module 10 according to a first preferred embodiment of the invention. Semiconductor laser module 10 comprises a tapered laser diode 12 and beam shaping optics in the form of a plano-convex spherical cylindrical lens 14. Lens 14 is arranged directly in front of the output facet 16 of the tapered laser diode 12.

[0040] The beam output by the tapered laser diode 12 is highly astigmatic and elliptical. The reason for this is that the laser diode facet has an elliptical cross section and has different divergences for the fast axis (fast axis divergence angle θ.sub.f between 24 and 46°) and the slow axis (the slow axis divergence angle θ.sub.s, depends on the taper angle θ.sub.t). With a taper angle of 6° the slow axis divergence angle θ.sub.s of the output beam amounts to about 20° due to diffraction at the facet 16. In the context of the example discussed with respect to FIGS. 1 to 3, the M.sup.2 factors are assumed to be approximately 1 both for the slow axis direction and the fast axis direction.

[0041] Tapered laser diodes are index-guided devices and typically have an active layer with a cross section size at the facet of about 0.1 μm by 150 μm. The 1/e.sup.2 diameters of the emission stripe are typically 1 μm and 190 μm directly at the facet, due to beam spreading into the surrounding lower index material, which leads to an ellipticity (ratio of minimum to maximum beam diameter) of less than 0.01. Different design parameters may result in ellipticity in the range from 0.002 to 0.03.

[0042] As best shown in FIGS. 2 and 3, the beam 18 is strongly astigmatic, since, in the direction perpendicular to the quantum well, it appears to diverge from a virtual source S.sub.1 in close proximity of the output facet 16, but, in the plane of the quantum well, it appears to diverge from a virtual source S.sub.2 that is located deeply inside the gain region 20. The magnitude of the astigmatism is the linear distance between S.sub.1 and S.sub.2. Whereas for index-guided, low power laser diodes the astigmatism amounts typically to 5 μm-15 μm and for gain-guided diodes the astigmatism may be as large as 50 μm, for tapered laser diodes the astigmatism may be larger than 500 μm or even more than 1000 μm in case of high power devices.

[0043] To correct the astigmatism, the plano-convex spherical cylindrical lens 14 is arranged such that the laser beam is circular (i.e. has an ellipticity of less than 0.87) at the axial position where the circular cylindrical surface 22 of the lens 14 intersects the beam axis 24 and configured such that the fast axis divergence of the laser beam 28 behind the lens 14 is reduced to match the divergence of the slow axis divergence of the laser beam 28 behind the lens.

[0044] There is a special condition, called aplanatic condition, in which a spherical surface (or, if only one transversal direction is considered, a circular cylindrical surface) introduces no spherical aberration. In the aplanatic condition not just all orders of spherical aberration, but also primary coma is zero for finite angles, which makes the lens more tolerant to misalignment. A spherical lens configured so as to satisfy the aplanatic condition may thus be considered more powerful than an aspheric lens, since the latter introduces primary coma. The aplanatic condition is mainly used to image a point source like in a microscope objective, but also works for line sources, which is a good approximation for the laser diode facet. FIG. 4 illustrates the aplanatic condition for a light ray originating from a source O in an optically thick medium with refractive index n and passing to an optically thinner medium with refractive index n′, i.e. n>n′. The inclination angle α between the optical axis and the light ray is reduced at the boundary surface between the two media, which is spherical (or circular cylindrical, as in the invention). For an observer behind the spherical surface, all the light rays originating from O thus appear to originate from the virtual image O′. When the aplanatic condition is not met, the position of the virtual image O′ depends on the inclination angle α, i.e. there is spherical aberration. No spherical aberration occurs when the aplanatic condition is met. This means, mathematically, that:

[00007] $\begin{matrix} \frac{\sin .Math. .Math. α^{'}}{n^{' .Math.}} = \frac{\sin .Math. .Math. α}{n} & (Eq . .Math. 8) \end{matrix}$

Furthermore, the distances I between the source O and the point of intersection P of the spherical (circular cylindrical) surface and the optical axis and I′ between the virtual image O′ and P have to satisfy the following relationships:

[00008] $\begin{matrix} l = \frac{n + n^{'}}{n} .Math. r .Math. .Math. and & (Eq . .Math. 9) \\ l^{'} = \frac{n + n^{'}}{n^{'}} .Math. r, & (Eq . .Math. 10) \end{matrix}$

where r denotes the radius of the curved surface.

[0045] A comprehensive deduction of the aplanatic condition can be found in P. Drude, “Lehrbuch der Optik, 3.sup.rd edition, ISBN: 3864547873 (Trapeza). When the optically thinner medium is air, n′=1, and the relationships simplify to:

[00009] $\begin{matrix} \sin .Math. .Math. α^{'} = \frac{\sin .Math. .Math. α}{n}, & (Eq . .Math. 11) \\ l = \frac{n + 1}{n} .Math. r .Math. .Math. and & (Eq . .Math. 12) \\ l^{'} = \frac{n + 1}{n^{' .Math.}} .Math. r . & (Eq . .Math. 13) \end{matrix}$

[0046] According to the illustrated embodiment of the invention, the plano-convex spherical cylindrical lens 14 is configured and arranged to satisfy the aplanatic condition in such a way that the virtual source S2 of the slow axis divergence substantially coincides with the virtual image of the virtual source S1 of the fast axis divergence.

[0047] In the theoretical situation of FIG. 4, there is only one boundary surface. In practice, however, the laser beam 18 enters the optically denser medium at the planar surface 26 of the lens 14 and leaves it across the circular cylindrical surface 22. As illustrated in FIG. 5, a light ray with inclination angle β is refracted at the first boundary surface such that:

[00010] $\begin{matrix} \sin .Math. .Math. β^{'} = \frac{n^{'}}{n} .Math. \sin .Math. .Math. β (Snell' .Math. s .Math. .Math. law), & (Eq . .Math. 14) \end{matrix}$

where β′ is the inclination angle of the beam inside the lens. If the aplanatic condition is satisfied, it follows for the refraction at the second, curved, boundary surface:

[00011] $\begin{matrix} \sin .Math. .Math. β^{″} = \frac{n^{'}}{n} .Math. \sin .Math. .Math. β^{'} = {(\frac{n^{'}}{n})}^{2} .Math. \sin .Math. .Math. β & (Eq . .Math. 15) \end{matrix}$

where β″ is the inclination angle of the beam after passage through the lens 14. What follows from this relation (with n′ being the refractive index of air, i.e, n′=1) is that for given fast and slow axis divergences (which depend on the laser diode configuration), the refractive index n of the lens 14 is chosen such that the equation

[00012] $\begin{matrix} n = \sqrt{\sin .Math. .Math. u_{f} / \sin .Math. .Math. (u_{s} .Math. \frac{M_{f}^{2}}{M_{s}^{2}})} = \sqrt{\sin .Math. .Math. u_{f} / \sin .Math. .Math. u_{s}} & (Eq . .Math. 16) \end{matrix}$

where u.sub.f denotes the uncorrected fast axis divergence half-angle (i.e, the fast axis divergence half-angle before the lens 14) and u.sub.s denotes the uncorrected slow axis divergence half-angle (i.e. the slow axis divergence half-angle before the lens 14), is satisfied. In practice, this means that the material of the lens 14 is chosen such that its refraction index satisfies

93%.Math.√{square root over (sin u.sub.f/sin u.sub.s)}≦n≦107%.Math.√{square root over (sin u.sub.f/sin u.sub.s)}. (Eq. 6′)

[0048] Those skilled will note that some spherical aberration results from the presence of the first, planar, surface 26 of the lens 14. In order to keep these spherical aberrations small, the planar surface 26 is positioned as close to the facet 16 as possible. In practice, this means that the distance Δ between the planar surface 26 and the facet 16 should not exceed 1 mm. In the illustrated embodiment, Δ is comprised in the range from 50 μm to 500 μm. The thickness of the lens 14 is then selected such that d.sub.f/d.sub.s=M.sub.f.sup.2/M.sub.s.sup.2, i.e. (in this example) such that the beam is circular, at the axial position where the circular cylindrical surface 22 of the lens 14 intersects with the beam axis 24. During the design of the beam shaping optics, this lo step is preferably carried out using a ray-tracing program or a lens design program on a computer.

[0049] Finally, the radius of curvature r of the lens is chosen such that, after passage of the laser beam through the lens, the fast axis divergence is substantially in agreement with the slow axis divergence (since in this example M.sub.f.sup.2≈1 and M.sub.s.sup.2≈1). With reference to FIG. 5, let h denote the 1/e.sup.2 radius of the beam at the axial position where the circular cylindrical surface 22 of the lens 14 intersects with the beam axis 24. Assuming that the lens 14 reduces the fast axis divergence to the slow axis divergence, it follows from l′=(n+1).Math.r that

[00013] $\begin{matrix} r = \frac{l^{'}}{n + 1} = \frac{h}{(n + 1) .Math. \tan .Math. .Math. u_{s}} = \frac{d_{f}}{2 .Math. (n + 1) .Math. \tan .Math. .Math. u_{s}}, & (Eq . .Math. 17) \end{matrix}$

where d.sub.f designates here the 1/e.sup.2 diameter of the beam at the axial position of the intersection of the curved lens surface with the beam axis 24. With some tolerance, one finally obtains:

[00014] $\begin{matrix} 82 .Math. % .Math. \frac{d_{f}}{2 .Math. (n + 1) .Math. \tan .Math. .Math. u_{s} .Math.} \leq r \leq 115 .Math. % .Math. \frac{d_{f}}{2 .Math. (n + 1) .Math. \tan .Math. .Math. u_{s}} . & (Eq . .Math. 7^{'}) \end{matrix}$

[0050] With the above parameters of the plano-convex spherical cylindrical lens 14, the beam 28 obtained after the lens 14 has substantially circular power density distribution and substantially equal beam divergence angles in fast and slow axis directions. With the above assumption M.sub.f.sup.2≈1 and M.sub.s.sup.2≈1, it is, therefore, comparatively easy to couple the obtained beam into an optical fiber with a good coupling efficiency. Indeed, the laser beam can be collimated using an aspheric lens or a lens group 30, as shown in FIGS. 2 and 3.

[0051] It shall be noted that the way of matching the fast and slow axis divergence angles may not be possible with every type of laser diode or amplifier diode, since a certain amount of astigmatism is required for a given pair of divergence angles. It is the inventors' merit to have recognized that the correction mechanism presented herein can be used for tapered laser diodes and tapered amplifier diodes, due to the comparably high value of the astigmatism.

[0052] As will be appreciated, the proposed scheme is a compact and powerful way to remove astigmatism and to circularize the beam of a tapered laser diode, which is highly astigmatic and elliptic. Only a single plano-convex spherical cylindrical lens with a specific refractive index is necessary for this purpose. An aspheric lens may thereafter be used to collimate the beam and/or to couple the beam into an optical fiber. The far field beam profile, or the focused spot, can be free of the side lobes and be nearly diffraction-limited. Due to the low number of lenses and the absence of truncating elements, the total power loss of the optical system may be reduced to less than 20%, which is much less than in systems using anamorphic prisms, a single mode fiber or a small aperture for circularizing a beam output by a laser or amplifier diode.

[0053] A very interesting advantage of using a plano-convex spherical cylindrical lens is that such lenses are easy to manufacture with high precision and also, compared with aspheric lenses, inexpensive. Since, in the illustrated configuration, spherical aberration is substantially eliminated, the use of a spherical cylindrical lens instead of an aspheric lens does not negatively affect beam quality. To the contrary, with primary coma being also eliminated, slight misalignment does not translate into a significantly deteriorated beam quality as in the case of an aspheric lens.

Example 1

[0054] A semiconductor laser module as shown in FIGS. 1 to 3 has been implemented with a tapered laser diode (wavelength λ=980 nm). The taper angle of the gain region was 6° (full angle). The facet measured 425 μm (slow axis direction) to 3 μm (fast axis direction). Without correction, fast axis divergence was 42° (full angle) and slow axis divergence 20° (full angle) at an output power of 8.5 W. M.sub.f.sup.2=M.sub.s.sup.2=1. For matching the fast axis divergence to the slow axis divergence, a fused silica plano-convex spherical cylindrical lens (n=1.45) with r=1.0 mm and a thickness of 1.42 mm (on the beam axis) was placed at a distance Δ=150 μm from the facet of the tapered laser diode. The conditions on n (Eq. 6) and r (Eq. 7) were thus met. Ellipticity of the laser beam on the curved surface of the lens amounted to 0.97 (i.e. the beam was circular). The fast axis divergence after the lens was reduced to 20.1° (full angle), which is approximately equal to the slow axis divergence.

Example 2

[0055] The identical laser diode of example 1 was chosen with the difference that the M.sub.f.sup.2=1 and M.sub.s.sup.2=2. For matching the fast axis divergence to M.sub.f.sup.2/M.sub.s.sup.2 times the slow axis divergence, a plano-convex spherical cylindrical lens made of P-SF68 (Schott, n=1.96) with r=0.51 mm and a thickness of 0.46 mm. (on the beam axis) was placed at a distance Δ=150 μm from the facet of the tapered laser diode. The conditions on n (Eq. 6) and r (Eq. 7) were thus met. Ellipticity of the laser beam on the curved surface of the lens amounted to 0.51, which corresponds to the desired ratio of d.sub.f/d.sub.s=M.sub.f.sup.2/M.sub.s.sup.2=0.5. The fast axis divergence after the lens was reduced to 10.1°, which is approximately equal to half the slow axis divergence.

[0056] FIG. 6 relates to a second preferred embodiment of the invention and shows a semiconductor laser diode array 32 comprising a stack of tapered laser diodes 34, each comprising a plano-convex spherical cylindrical microlens 36 arranged in front of its facet 38.

[0057] Each tapered laser diode 34 produces a highly astigmatic and elliptical beam at its output. It is assumed that M.sub.f.sup.2≈1, whereas M.sub.s.sup.2 is larger (e.g. M.sub.s.sup.2 comprised in the range from 2 to 5), as if often the case in industrial systems because this makes laser diodes more robust against back-reflections. Each lens 36 is positioned such that its circular cylindrical surface intersects the respective beam axis at an axial position where the ratio d.sub.s/d.sub.f of the slow axis beam diameter d.sub.s of the laser beam (after passage through the lens 36) to the fast axis beam diameter d.sub.f of the laser beam (after passage through the lens 36) amounts to a value comprised in the range from 0.87.Math.M.sub.s.sup.2/M.sub.f.sup.2 to 1.15 M.sub.s.sup.2/M.sub.f.sup.2. Furthermore, the parameters of each lens are chosen such that the fast axis divergence of the respective laser beam is reduced to approximately match M.sub.f.sup.2/M.sub.s.sup.2 times the slow axis divergence. As a consequence, by focusing the beam, one obtains a circular power density distribution.

[0058] As in the previously discussed embodiment, the material of each lens 36 is chosen such that its refraction index satisfies Eq. 6. The distance Δ′ between the planar surface of each lens 36 and the respective facet 38 is comprised in the range from 50 μm to 500 μm. The radius of curvature of each lens is chosen such that, after passage of the laser beam there through, the fast axis divergence is substantially in agreement with the slow axis divergence multiplied with M.sub.f.sup.2/M.sub.s.sup.2, i.e. Eq. 7 is satisfied.

Example 3

[0059] The tapered laser diode of example 2 can be used to build a laser diode stack, which consists of 20 laser diode bars, each of which consists of 10 tapered laser diodes with a pitch (center to center of the tapered laser diode) of 1 mm. The spacing between the laser diode bars may be 0.5 mm, such that the laser diode stack has a square emitting area of 10 mm by 10 mm. Since each emitter is capable of producing at least 5 W of optical power, the stack can deliver more than 1000 W of optical power. The optics detailed in example 2 may be mounted in front of each laser diode bar and afterwards a multi lens array can collimate each tapered laser diode individually, each facet of the multi lens array comprising a lens with the geometrical ratio equal to d.sub.f/d.sub.s. After the multi lens array, a collimated beam consisting of 200 smaller beams is obtained, which can be focused by a lens into a fiber of 100 μm diameter with a numerical aperture of 0.22.

[0060] FIG. 7 shows a semiconductor laser module 40 according to a further preferred embodiment of the invention. Semiconductor laser module 40 comprises tapered amplifier diode 42 and beam shaping optics in the form of a plano-convex spherical cylindrical lens 44 arranged directly in front of the output facet 46 of the tapered laser diode 42. The tapered gain region 48 amplifies a seed beam 50 input at input facet 52 of the tapered amplifier diode 42. The plano-convex spherical cylindrical lens 44 is configured and positioned so as to match the fast axis divergence θ.sub.f with θ.sub.s.Math.M.sub.f.sup.2/M.sub.s.sup.2 at the position where d.sub.f/d.sub.s=M.sub.f.sup.2/M.sub.s.sup.2, in accordance with what has been explained with respect to FIGS. 1 to 3.

Example 4

[0061] A semiconductor laser module has been implemented with a tapered laser diode (wavelength λ=980 nm). The taper angle of the gain region was 6° (full angle). The facet measured 425 μm (slow axis direction) to 3 μm (fast axis direction). Without correction, fast axis divergence was 42° (full angle) and slow axis divergence 20° (full angle) at an output power of 8.5 W with M.sub.f.sup.2=M.sub.s.sup.2=1. For beam circularization and collimation of both axes, a plano-convex spherical cylindrical lens with r=1.55 mm and a thickness of 2.45 mm (on the beam axis) was placed at a distance Δ=250 μm from the facet of the tapered laser diode. The lens was made from gradient-index material with a rotationally invariant quadratic refractive index variation as shown in FIG. 8. The conditions of Eq. 6 and Eq. 7 are thus met for the refractive index n.sub.0=1.62 at the center of the optics (on the beam axis) and for the radius of curvature. The ellipticity of the laser beam on the curved surface of the lens amounted to 0.97 (i.e. the beam was circular). The fast axis divergence after the lens was reduced to less than 0.1°, which was approximately equal to the slow axis divergence obtained by the refractive-index variation. The beam thus could be considered collimated after passing the lens. The beam diameter directly behind the lens amounted to 0.9 mm.

[0062] FIGS. 9 and 10 illustrate the beam shaping optics of a semiconductor laser module according to a preferred embodiment. As in the embodiment of FIGS. 1, 2 and 4, the semiconductor laser module comprises a tapered laser diode, which, however not shown. The beam shaping optics comprise a plano-convex spherical cylindrical GRIN lens 114 arranged directly in front of the output facet 116 of the tapered laser diode.

[0063] The beam output by the tapered laser diode is strongly astigmatic. The astigmatism is corrected and the beam is collimated by the plano-convex spherical cylindrical GRIN lens 114. The GRIN lens 114 is arranged such that the laser beam is circular (i.e. has an ellipticity of less than 0.87) at the axial position where the circular cylindrical surface 122 of the lens 114 intersects the beam axis 124. The refractive index of the GRIN lens 114 varies in the direction of the cylinder axis 123 (parallel to the slow axis direction) and in the fast axis direction according to a parabolic function having its maximum n.sub.0 on the beam axis. In the direction of the beam axis, the refractive index of the GRIN lens 114 is substantially constant. n.sub.0 is selected so as to satisfy the condition of Eq. 6. The radius of curvature of the circular cylindrical surface 122 is chosen so as to satisfy the condition of Eq. 7 (after substitution of n by n.sub.0). The GRIN lens reduces the divergence in the fast axis direction and in the slow axis direction in such a way as to make the divergences match and to obtain a collimated beam 125. The collimated beam 125 can be easily focused with a spherical lens 126.

[0064] While specific embodiments have been described in detail, those with skill in the art will appreciate that various modifications and alternatives to those details could be developed in light of the overall teachings of the disclosure. Accordingly, the particular arrangements disclosed are meant to be illustrative only and not limiting as to the scope of the invention, which is to be given the full breadth of the appended claims and any and all equivalents thereof.

Semiconductor Laser Module

Inventors

Cpc classification

Classification Explorer

H01S5/22

ELECTRICITY

Classification Explorer

H01S5/02251

ELECTRICITY

Classification Explorer

H01S5/005

ELECTRICITY

Classification Explorer

G02B3/06

PHYSICS

Classification Explorer

G02B19/0057

PHYSICS

Classification Explorer

G02B3/0087

PHYSICS

Classification Explorer

G02B19/0014

PHYSICS

Classification Explorer

H01S5/1053

ELECTRICITY

Classification Explorer

G02B27/0994

PHYSICS

Classification Explorer

H01S5/405

ELECTRICITY

Classification Explorer

H01S2301/18

ELECTRICITY

Classification Explorer

H01S5/1014

ELECTRICITY

Classification Explorer

G02B27/0966

PHYSICS

International classification

Classification Explorer

G02B27/09

PHYSICS

Classification Explorer

G02B19/00

PHYSICS

Classification Explorer

H01S5/10

ELECTRICITY

Abstract

Claims

Description