Large optical cavity (LOC) laser diode with quantum well offset and efficient single-mode lasing along the fast axis

Abstract

The laser diode is configured to suppress lasing of first and higher order modes along a fast axis of a light beam emitted by the laser diode. An optical cavity is defined by a p-side of the laser diode, an n-side of the laser diode, and an active region located between the p-side and the n-side. The n-side has an n-waveguide layer forming at least a portion of a waveguide having a quantum well offset toward the p-side. According to some embodiments, the double cladding outcouples higher order modes. According to other embodiments, the double waveguide (e.g., symmetric and asymmetric) reduces the gain applied to the higher order modes.

Description

[0001] Related Applications

[0002] This application claims the benefit of priority to U.S. Provisional Patent Application No. 62 / 677,748, filed on May 30, 2018, which is incorporated herein by reference. Technical Field

[0003] The present disclosure relates to a laser diode. Background Art

[0004] Some laser diodes, such as those included in high power 8xx- and 9xx-nm commercial laser diode packages available from applicant nLIGHT Corporation, include a large optical cavity (LOC) for high power and high efficiency performance. Summary of the invention

[0005] Several embodiments of LOC laser diode structures are described. These embodiments have relatively thin p-waveguide thicknesses and also provide single-mode lasing operation in the vertical direction at high power and high efficiency. High power laser diodes are desirable for power scaling and lowering price (measured in cost per watt). High efficiency laser diodes are desirable for lowering energy consumption and extending device lifetime.

[0006] In some examples, the laser diode is configured to suppress lasing of first and higher order modes along a fast axis of a light beam emitted by the laser diode. The laser diode has an optical cavity defined by a p-side of the laser diode, an n-side of the laser diode, and an active region located between the p-side and the n-side, the n-side including an n-waveguide layer forming at least a portion of a waveguide having a quantum well offset to the p-side; and adjacent claddings on one or both of the p-side and the n-side, the adjacent claddings including outer and inner claddings, the inner cladding being located between and adjacent to the waveguide and the outer cladding, the outer and inner claddings having first and second refractive indices, respectively, the first refractive index of the outer cladding being greater than the second refractive index of the inner cladding and greater than the effective refractive index of the first order mode so as to couple it out of the waveguide. In some examples, a graded or stepped refractive index profile can be adjacent to the active region.

[0007] In other examples, the laser diode is configured to suppress lasing of first and higher order modes along a fast axis of a light beam emitted by the laser diode. The laser diode has an optical cavity defined by a p-side of the laser diode, an n-side of the laser diode, and an active region between the p-side and the n-side; and a set of n-waveguide layers on the n-side that form at least a portion of a dual waveguide, the set of n-waveguide layers including outer and inner n-waveguide layers to reduce optical gain applied to the first and higher order modes. The waveguide is symmetric in some examples and asymmetric (with respect to refractive index) in other examples.

[0008] Other aspects and advantages will become apparent from the following detailed description of embodiments which proceeds with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] FIG. 1 is a block diagram of a conventional laser diode according to the prior art.

[0010] Figure 2 is an annotated block diagram of an epitaxial layer stack including a cladding (also called a cladding), a waveguide, and an active layer arranged along a vertical axis according to a first single-mode laser diode design example (specifically, Design 1a) with a thin p-waveguide thickness (i.e., a large subwell offset) and a double n-cladding structure.

[0011] Figure 3 is a graph of the refractive index profile (RIP) of Design 1a showing a model of the light intensity distribution of the fundamental and first order modes, respectively, with effective refractive indices represented by dashed lines, for comparison with Figure 2 The refractive indices of the inner and outer n-claddings of the double-cladding example introduced in are compared.

[0012] Figure 4 is a graph of the RIP curves of single n-cladding design variants with a small refractive index step difference (Δn), and shows a model of the light intensity distribution of the fundamental mode of the variants and design 1a.

[0013] Figure 5 is a RIP plot for a single n-cladding design variant with large Δn and shows a model of the light intensity distribution of the fundamental and first-order modes, both of which are confined in the waveguide and thus lead to stimulated emission of multiple optical modes.

[0014] Figure 6 is an annotated graph of the RIP for a single-mode laser diode design 1b with extremely large (ie, essentially zero p-waveguide thickness) quantum well offset.

[0015] Figure 7 is an annotated plot of the RIP for a single-mode laser diode design with a double p-cladding structure and a thin p-waveguide thickness (ie, a large subwell offset).

[0016] Figure 8 are annotated graphs of two RIPs of the second design example, including two corresponding large number of subwell offset designs 2a and 2b with step and graded index RIP layers, respectively.

[0017] Fig. 9 is an annotated graph showing the logarithmic scale for Figure 8 Vertical far-field intensity distribution models for Designs 2a and 2b, which show the relatively small optical power present in the high-angle sidelobes.

[0018] Fig.10 are annotated graphs of two RIPs of the second design example, including two corresponding very large quantum well offset designs 2c and 2d with stepped and asymmetric graded-index RIP layers, respectively.

[0019] Fig.11 is an annotated graph showing the logarithmic scale for Fig.10 Vertical far-field intensity distribution models of designs 2c and 2d, which show the relatively small optical power present in the high-angle sidelobes.

[0020] Fig.12 is an annotated graph of the RIP of the third large quantum well offset design example, where the outer n-waveguide has a lower refractive index than the inner n-waveguide and where the optical confinement factor of the fundamental mode is enhanced to 0.5% while the optical confinement factor of the first-order mode is reduced to 0.3%.

[0021] Fig.13 is an annotated graph of the RIP of the fourth large quantum well offset design example, where a thin cladding separates the dual waveguide structure and where the optical confinement factor of the fundamental mode is enhanced to 0.4% while the optical confinement factor of the first-order mode is reduced to 0.2%.

[0022] Fig.14A is a block diagram of a laser diode assembly including a laser diode constructed according to one or more disclosed design examples.

[0023] Fig. 14B It is along Fig.14A Side view of the stepped laser diode base at line BB.

[0024] Fig.15 It is for traditional LOC design and has Figure 3 and 6 Graphs of simulated laser diode voltage (V) as a function of drive current (Amperes, A) for several thinner p-waveguide designs of the line of designs 1a and 1b shown in FIG.

[0025] Fig.16 and 17Plots of simulated electron and hole concentrations and conduction band edge under high-power operation for conventional LOC laser diode design and design 1a, respectively.

[0026] Fig.18 is a graph of measured voltage as a function of current, which indicates that for designs 1a and 1b Figure 3 and 6 The lines shown in FIG. 5 show several thin p-waveguide designs with progressively thinner p-waveguides, with the power consumption being smaller as the p-waveguide thickness decreases.

[0027] Fig.19 yes Fig.18 Plot of the measured fast-axis far field of the thinnest p-waveguide design, showing single-mode output along the fast axis for measured currents between 10A and 25A.

[0028] Fig. 20 2a is a graph of measured voltage and output power as a function of current for a conventional laser diode design and an improved design employing a step-index RIP layer along the lines shown in Design 2a.

[0029] Fig.21 It is a traditional laser diode design and Fig. 20 Plot of the measured efficiency as a function of current for the improved design.

[0030] Fig. 22 yes Fig. 20 Plot of the measured fast-axis far field as a function of angle for the improved design.

[0031] Fig.23 is an annotated graph showing the RIP of a combination of dual waveguides and dual cladding according to another embodiment. DETAILED DESCRIPTION

[0032] Laser diodes can be manufactured to operate over a wide range of wavelengths, depending on the materials selected. For example, such laser diodes can be made of InGaN, AlGaInP, GaAlAs, InGaAs, GaAsP, InGaAsP, InGaAsNSb, or GaInAsSb for laser emission wavelengths ranging from at least about 400 nm (InGaN) to about 3.4 μm (GaInAsSb). The laser emission wavelength refers to the actual emission wavelength of the laser diode. The actual emission wavelength varies with temperature, drive current, and laser cavity length; as used herein, the laser emission wavelength also refers to the spectral bandwidth that a laser diode can emit, which is typically measured in terms of the full width at half maximum or the full width 1 / e 2 quantify the intensity.

[0033] For example, FIG. 1 shows a vertical waveguide 100 of a conventional laser diode 112 that emits laser radiation 114 from a front facet 118 that is longer in one direction than in an orthogonal direction, such that the emitted laser radiation 114 has a smaller divergence in a direction parallel to the longer facet dimension and a larger divergence in a direction parallel to the smaller facet dimension. The axis parallel to and along the longer facet dimension is referred to as the slow axis 120; the axis parallel to and along the smaller facet dimension is referred to as the fast axis 124. Also, as shown in FIG. 1 , the vertical waveguide 100 includes a p-waveguide 130 on the p-side 132 of the pn junction, an n-waveguide 134 on the n-side 136 of the pn junction, and an active layer 138 (also referred to as an active region, and the word layer is sometimes omitted when describing portions of a stack) having one or more quantum well layers and other portions to produce optical gain. The n and p waveguide layers and the layers in between are collectively referred to as waveguides (i.e., the vertical waveguide 100 in FIG. 1 ).

[0034] The active region of a laser diode is in the intrinsic region, into which carriers (electrons and holes) are pumped from the n- and p-regions, respectively (i.e., the n- and p-doped materials on opposite sides of the pn junction). The term "active region" is also often used to refer to the portion of the laser diode that generates optical gain. In quantum well lasers, the quantum well layers serve as the active region. Therefore, in the following examples, the so-called quantum well laser diode includes an active region based on quantum wells.

[0035] In a conventional laser diode 112, the p-waveguide 130 constitutes a significant portion of the total thickness of the vertical waveguide 100. In other words, the quantum well located between the p-side 132 and the n-side 136 is located more or less in the middle of the vertical waveguide 100. However, as P. Crump et al., "Efficient High-Power Laser Diodes," IEEE Journal of Selected Topics in Quantum Electronics, Vol. 19, No. 4, 1501211, July-August 2013, most of the voltage and light losses occurring at high current 140 are attributed to the thickness of the p-waveguide 130.

[0036] Some laser diode designs attempt to reduce thickness, which results in a quantum well offset toward the thinner p-side perpendicular to the waveguide. The large quantum well offset reduces the electrical series resistance attributable to the p-waveguide, reduces optical losses and leakage currents associated with carrier accumulation in the p-waveguide under large forward biases, and thus results in improved power and efficiency. However, the large quantum well offset toward the p-side of the waveguide in the LOC structure results in the appearance of higher order modes in the perpendicular direction. This is particularly true for the onset of the first order mode, which deteriorates the beam quality in the perpendicular direction and reduces the power and efficiency of the diode.

[0037] P.Crump et al. describe such a design, which has a reduced p-waveguide thickness and attempts to produce single-mode emission in the vertical direction. Specifically, for example, P.Crump et al. Fig.10 The design shown in (b) has a reduced p-waveguide thickness and a reduced refractive index step difference (Δn=0.02) between the waveguide and the n-cladding in order to preferentially outcouple the first (higher) order optical mode. This results in single-mode emission in the perpendicular direction, but introduces certain drawbacks as understood by the inventors. For example, the fundamental mode extends significantly into the cladding, resulting in a large, ill-defined near field, reduced beam quality, and higher intrinsic losses from overlap with the cladding. Moreover, the small energy barrier provided by the n-cladding as a result of the relatively small Δn (i.e., the n-cladding refractive index is only slightly lower than that of the waveguide) may not be sufficient to block carrier / hole leakage into the cladding, which can reduce the intrinsic efficiency of the laser diode due to carrier leakage. In other words, there is a value of Δn below which the waveguide supports one confined mode (i.e., a small Δn for single-mode lasing in the fundamental mode) and above which the waveguide begins to confine the first-order mode (i.e., a large Δn that causes multi-mode lasing including the first-order mode).

[0038] Design Example 1: Double Cladding Structure

[0039] Figure 2 A laser diode 200 is shown that represents a design 1a featuring a dual (i.e., inner and outer) n-cladding structure. The laser diode 200 includes an active region 210 located between a p-waveguide 214 and an n-waveguide 216. The n-waveguide 216 is between the active region 210 and the inner n-cladding 220. Additionally, the outer n-cladding 230 is located between the inner n-cladding 220 and the n-substrate 240 at the bottom of the epitaxial layer stack. A p-cap 250 is located on top of the p-cladding 254 between the p-cap 250 and the p-waveguide 214. Note that although this and other examples are described with reference to directions indicated as "above," "below," "upper," "lower," "vertical," "horizontal," etc., these terms are used for ease of description and do not imply any particular spatial orientation or limitation. Furthermore, in these graphs, the horizontal arrows clarify which vertical axis corresponds to which plot.

[0040] Figure 3A graph 300 of design 1a is shown that includes a double n-cladding structure 310 having an outer n-cladding 312 and a relatively thin inner n-cladding 316. Between the n and p waveguides 320, 326 (the latter adjacent to the p-cladding 338) is an active region 340 that includes a quantum well 344. The n and p waveguides 320, 326 and the active region 340 together form a waveguide 350. For relatively small far-field divergence angles, the overall thickness of the waveguide 350 remains large (as in conventional LOC designs). With such a large optical cavity, multiple optical modes in the vertical direction can be supported.

[0041] Due to the relatively thin p-waveguide 326, the quantum well 344 is significantly offset to the p-side of the waveguide 350, which creates a large quantum well offset that improves power and efficiency. However, without the outer n-cladding 312 used to suppress the undesired modes for reasons explained below, the large quantum well offset will cause the confinement factor of the first-order mode (Γ, measured as the overlap of the optical mode with the quantum well) to be greater than or equal to that of the fundamental mode - lasing in both modes will then degrade the laser diode performance. But if Figure 3 As shown, the refractive index of the outer n-cladding 312 is selected so that it is both lower than the effective refractive index 352 of the fundamental mode 356 and higher than the effective refractive index 360 of the first-order mode 370. The refractive index of the outer n-cladding 312 is also slightly lower than the refractive index of the waveguide 350. The advantages of the inner n-cladding 316 will be discussed later. Figure 4 and 5 As discussed above, the refractive index of the waveguide 350 and the outer n-cladding 312 is substantially lower than that of the waveguide 350 and the outer n-cladding 312. Such an arrangement ensures that the fundamental mode 356 is a guided mode of the waveguide 350, while the undesired mode 370 is an anti-guided mode (i.e., a leaky mode) with high optical loss. The fundamental mode 356 is well confined by the waveguide 350, and the first-order (and higher-order) modes 370 are outcoupled to the cladding with high loss, so that the laser diode of Design 1a will emit in single-mode laser in the vertical direction.

[0042] The effective refractive indexes 352 and 360 help illustrate how the double n-cladding structure defines the fundamental mode and rejects the first (and higher) order modes. In the context of waveguide optics, the term effective refractive index is generally understood to refer to the refractive index associated with the propagation of the waveguide mode. The effective refractive index depends not only on the wavelength, but also (for multimode waveguides) on the mode in which the light propagates. Therefore, it is also called the mode refractive index. For example, the effective refractive index can be obtained by numerical mode calculation.

[0043] Figure 4 is a graph 400 comparing the fundamental mode 356 of design 1a with the fundamental mode 402 of another design in which there is a relatively small Δn 404 between the n-waveguide 406 and the single n-cladding 408. As described above, because Figure 4 The so-called small Δn design has no inner n-cladding layer 316 ( Figure 3 ), so its fundamental mode 402 extends 414 significantly into the single n-cladding 408 (compared to the fundamental mode 356), resulting in a large near-field, reduced beam quality, and higher intrinsic losses due to overlap with the n-cladding 408. On the other hand, Figure 5 As shown in the graph 500 of , a large Δn design will result in laser emission of higher order modes confined to the waveguide 550. Therefore, to overcome the drawbacks of the small and large Δn designs, the design 1a has a high bandgap energy barrier provided by the inner n-cladding 316, which suppresses the leakage of carriers (holes) into the n-cladding 310, thereby suppressing higher order modes while ensuring the high intrinsic efficiency of the double n-cladding laser diode design.

[0044] Figure 6 is a graph 600 showing design 1b with extreme quantum shift. As in design 1a, a double n-cladding structure 610 is employed. Compared to design 1a, design 1b has an extremely thin or no p-waveguide 620. For reasons similar to those discussed previously, the laser diode of design 1b will also laser in single mode. Both double n-cladding designs 1a and 1b (generally referred to as design 1) provide sufficient confinement of the fundamental mode for spreading smaller modes into the cladding and well-defined modal near fields, while suppressing higher order modes by introducing sufficient losses.

[0045] The following table provides example thicknesses and refractive indices for Designs 1a and 1b.

[0046]

[0047]

[0048] The skilled artisan will appreciate that the actual layer thicknesses are typically independently selected to provide the desired optical coupling. Some or all layer thicknesses may be adjusted as desired. Typically, the inner n-cladding layer 316, including its refractive index and layer thickness selection, is selected to have a much lower refractive index than the waveguide, and typically also has a much higher bandgap energy than the waveguide.

[0049] Figure 7 is a graph 700 showing another dual cladding design on the p-side instead of the n-side. The dual p-cladding structure 710 will result in greater electron leakage into the p-cladding and significant mode loss from the p-cladding, but this design still achieves similar mode discrimination to suppress higher order modes due to the position and refractive index of the outer p-cladding 712 and inner p-cladding 716, which are functionally similar to the outer and inner n-claddings 312, 316.

[0050] Design Example 2: Active Region with High-Index RIP

[0051] In Design Example 1, the confinement factor is reduced compared to conventional LOC laser diodes due to the large quantum well offset. For example, the confinement factors for Design 1a and Design 1b are 0.26% and <0.1%, respectively, compared to 0.5-0.6% in conventional LOC designs. This results in lower modal gain and increased threshold current, and may limit the power and efficiency of the laser diode in some applications.

[0052] Design Example 2 (including variants 2a, 2b, 2c, and 2d) has a similar double n-cladding structure as Design 1, but with a high-index RIP adjacent to the active region that enhances optical confinement at the quantum wells, thus providing higher modal gain and reduced threshold current. Note that Figure 8 In , Design 2a is shown as having a lower overall refractive index than Design 2b, but this difference is simple so that one design does not overshadow the other - the actual refractive indices may be similar. Likewise, Fig.10 Similar overall differences are shown between designs 2c and 2d, which are also provided for clarity.

[0053] Figure 8 is a graph 800 showing Design 2a and Design 2b. In both designs, the high-index RIP surrounds the quantum well 844 to help increase mode overlap with the quantum well and has a higher refractive index than that of the waveguide 850. In other words, the high-index RIP acts as a higher-index layer to help increase the confinement factor.

[0054] In Design 2a, the high-index RIP is a step-index layer 862. In Design 2b, a gradient-index RIP 870 is used. The confinement factor for both designs is 0.52%, which is about twice that of Design 1a. Other possible designs include: a design similar to Design 2a, but with an asymmetric high-index step-index RIP around the active region, i.e., the step-index layer on the p-side is thinner than the step-index layer on the n-side or has no thickness; and a design similar to Design 2b, but with a faster index gradient or no thickness on the p-side of the gradient-index RIP.

[0055] Fig. 9 Graph 900 shows that the RIP of Designs 2a and 2b results in high angle sidelobes in the vertical far field, but these sidelobes contain only a small portion of the total power. For example, for both designs, about 1% of the power is enclosed in the sidelobes. In other words, assuming power loss at high far field angles, the sidelobes account for about 1% of the power loss for Design 1a.

[0056] Fig.10Graph 1000 shows the RIP of designs 2c and 2d, each with extreme quantum well offsets. Because the confinement factor in the designs with extreme quantum well offsets is low, such as the confinement factor in design 1b (<0.1%), the RIP of designs 2c and 2d is Fig.10 The refractive index distribution around the quantum well 1044 is modified.

[0057] Design 2c includes a high index step-index RIP layer 1062 on either side of the quantum well 1044, similar to the previously described design 2a. Design 2c has a confinement factor of 0.26%, which is the same as that of design 1a, even with the p-waveguide completely removed. Design 2c can also have an asymmetric step-index RIP around the active region, i.e., the step-index layer on the p-side is thinner than or has no thickness on the n-side.

[0058] Design 2d includes an asymmetric graded index RIP layer 1070 with a faster index gradient or no thickness on the p-side. The purpose of the asymmetric RIP layer is to improve the confinement factor while minimizing the p-waveguide thickness. Specifically, in Design 2d, a confinement factor of 0.38% is achieved in the extreme quantum well offset design, which is a significant improvement over Design 2c. However, in other designs, a symmetric graded index RIP layer (not shown) or other types of RIP layers may be used to improve the confinement factor.

[0059] Fig.11 Graph 1100 plots the modeled vertical far-field intensity distributions for Designs 2c and 2d. Both designs have small power losses in the high-angle sidelobes, with Design 2c having less than 1% power loss and Design 2d having less than 2% power loss.

[0060] Example thicknesses and refractive indices according to variations of this Design Example 2 would be similar to those of Tables 1 and 2 except for the thin high index RIP surrounding the quantum wells.

[0061] Design Example 3: Dual Waveguide Structure with Asymmetric Refractive Index

[0062] Fig.12 is a graph 1200 showing the RIP of Design Example 3. The design includes a dual waveguide 1250 including an outer n-waveguide layer 1260 , an inner n-waveguide layer 1262 , and a large quantum well offset to a thin p-waveguide 1266 on the p-side of the waveguide 1250 .

[0063] The outer n waveguide 1260 has a slightly lower refractive index than the inner n waveguide 1262 and the thin p waveguide 1266. Therefore, the dual waveguide 1250 is asymmetric with respect to the refractive index because they are different on opposite sides of the waveguide 1250. As explained in the following paragraphs, there is effective power of the fundamental mode in the outer n waveguide 1260. Therefore, the outer n waveguide layer 1260 is not referred to as an inner n cladding. However, whether it is referred to as an inner n cladding or an outer n waveguide is largely irrelevant to the principle of operation: in contrast to Design Examples 1 and 2, both the fundamental and higher order (i.e., first order) modes are confined modes in Design Example 3. In Design Example 3, the first order mode is not confined because it is outcoupled, but because its modal gain is lower than the modal gain of the fundamental mode.

[0064] Under the dual waveguide 1250, the fundamental mode 1256 tends to overlap more with the higher refractive index region of the inner n-waveguide 1262, while the first-order mode 1270 overlaps more with the lower refractive index region of the outer n-waveguide 1260. The fundamental mode 1256 of the specific structure has an enhanced confinement factor of 0.5% compared to 0.2% under the symmetric waveguide of the same thickness and quantum well offset. The first-order mode 1270 has a confinement factor of 0.3%, which is significantly reduced from 0.4% in the symmetric waveguide. The enhanced mode discrimination of the fundamental mode 1256 is generated by its increased mode gain relative to the first-order mode 1270, allowing the laser diode to operate in single mode in the vertical direction. For comparison, design examples 1 and 2 achieve enhanced mode discrimination of the fundamental mode by increasing the loss of the first-order mode.

[0065] The following table provides example thicknesses and refractive indices for Design 3.

[0066]

[0067] Design Example 4: Dual Waveguide Structure with Symmetrical Refractive Index

[0068] Fig.13 is a graph 1300 showing the RIP of Design Example 4. The design includes a dual waveguide 1350 formed of two waveguides separated by a thin cladding 1354, which is offset more toward the thinner waveguide 1360 than the thicker waveguide 1362. The thicker waveguide 1362 includes an n-waveguide and a p-waveguide 1366 located to the right of the thin cladding 1354. The thin cladding 1354 has a lower refractive index (and typically also a higher band gap energy) than the dual waveguide 1350. The quantum well is offset to the p-side of the dual waveguide 1350 with a very thin p-waveguide 1366 thickness.

[0069] In the case of a dual waveguide with unequal thickness in the two waveguides, the first order mode 1370 overlaps more with the thinner waveguide 1360 to the n-side of the thin cladding 1354, while the fundamental mode 1356 tends to overlap more with the thicker waveguide 1362 to the p-side of the cladding. Fig.13 With the particular structure of FIG. 1 , the fundamental mode 1356 has an enhanced confinement factor of 0.4% and the first-order mode 1370 has a confinement factor of 0.2%, resulting in improved mode discrimination between modes and single-mode operation in the perpendicular direction.

[0070] To characterize the above improvement according to one example, note that in a case with a diameter less than 0.9 cm –1 In laser structures with intrinsic mode losses, a mode discrimination factor that produces greater than two times (or more) the threshold (explained below) is considered good mode discrimination for single-mode operation. At this threshold, higher order modes are considered to be suppressed, which is a practical matter. However, those skilled in the art will appreciate that other thresholds may be suitable for different applications in order to increase the threshold current of the higher order modes to a desired amount above the threshold current of the fundamental mode. In practice, the threshold current is related to the limiting factor (modal gain) as well as the intrinsic mode losses (different modes have different losses), so the threshold current is not solely dependent on the limiting factor. Higher order modes with threshold currents greater than about two or three times the threshold current of the fundamental mode will not laser emit at all, that is, such a range is implementation specific and does not need to correspond directly to the limiting factor.

[0071] The thickness and refractive index of the cladding 1354 are carefully selected to achieve maximum mode discrimination between the fundamental and first order modes - a layer that is too thin will reduce the mode discrimination, but a layer that is too thick will result in modes confined in separate waveguides, which deviates from the LOC design principle. In other words, if the cladding 1354 is too thick, then this will result in two uncoupled waveguides, where light is confined to one waveguide with the quantum well, making the dual waveguide design paradigm inoperable.

[0072] The following table provides example thicknesses and refractive indices for Design 4.

[0073]

[0074] Laser Diode Components

[0075] The foregoing design examples are particularly suitable for fiber-coupled pump products, such as those available from applicant nLIGHT, Inc. of Vancouver, Washington. and pearl TMLaser diodes. For example, the ability to scale power by increasing the output power of each single emitter diode with high efficiency and without modifying the pump design or parts count significantly reduces pump cost (cost per watt). Moreover, the improved efficiency helps to reduce energy consumption and extend the life of laser diodes and laser diode products that incorporate such devices.

[0076] refer to Fig.14A , the laser diode assembly 1400 is described with reference to a right-handed xyz coordinate system 1480, wherein the z-axis extends upwardly out of Fig.14A plane. The laser diode assembly 1400 includes coherent laser diode groups 1402-1405, which are arranged to emit laser beams to corresponding phase masks, reflectors, and fast and slow axis collimators 1412-1415. For example, the laser diodes of group 1402 emit a beam in the x-axis direction, which is then redirected by the corresponding reflector of group 1412 to propagate in the y-axis direction. The laser diodes of each group are shifted or offset along the z-axis from the other laser diodes of the same group, and the associated reflectors are positioned so that the laser beam from the group is not blocked by these reflectors. As Fig. 14B As shown, the group of laser diodes 1402 is fixed to a stepped base 1420 to provide the appropriate z-axis offset; similar bases are provided for the remaining groups of laser diodes. For convenience, the beam propagation axes 1422-1425 of the bottom-most laser diode of each group are shown; the beam propagation axes of the remaining laser diodes of each group are similar, but offset along the z-axis.

[0077] The laser beams from the group of laser diodes 1402 are directed to a half-wave retarder 1430 and then combined with the laser beams from the group of laser diodes 1403 at a polarization beam splitter 1432, such that a vertically stacked group of beams is incident to a beam spacing compressor 1440, which may be provided as a pair of cylindrical lenses. The laser beams from the group of laser diodes 1405 are directed to a half-wave retarder 1434 by the group of reflectors 1415 and then combined with the laser beams from the group of laser diodes 1404 at a polarization beam splitter 1435 while being redirected by the group of reflectors 1414, such that a vertically stacked group of beams is incident to the beam compressor 1440.

[0078] Experimental and Simulation Results: Design Examples 1 and 2

[0079] The disclosed LOC laser diodes have improved power and efficiency due to lower operating voltage from reduced series resistance and higher slope efficiency (i.e., slope efficiency of power / current from the linear portion of the laser power characteristic above the threshold current) from lower optical losses. The laser diode performance of Design Examples 1 and 2 has been simulated using a self-consistent physics-based laser diode simulation tool. For comparison, simulation results for a conventional LOC laser diode are also provided.

[0080] Fig.15 The simulated laser diode voltage as a function of drive current is plotted for a conventional LOC design and several thin p-waveguide designs following the double-cladding thin p-waveguide example of Design Example 1. Most of the resistance in a conventional LOC comes from the relatively thick p-waveguide due to the low mobility of holes. Fig.15 As shown, making the p-waveguide thinner significantly reduces the series resistance and therefore reduces the voltage in the laser diode.

[0081] In conventional LOCs, bias-driven carrier leakage into the p-waveguide is the major power and efficiency detractor, especially under high-power and high-current operation. Fig.16 Simulated electron and hole concentrations and conduction band edges for a conventional LOC laser diode design under high power operation are shown. Band bending occurs (conduction band edge in the figure), leading to increased carrier leakage and accumulation in the p-waveguide (both electrons and holes, as shown in the figure). This causes enhanced optical losses through free carrier absorption in the p-waveguide (especially intervalence band absorption), and manifests as power saturation with increasing current due to significant carrier accumulation at high injection currents. Reducing the p-waveguide in a LOC structure such as in Design Example 1 is shown to mitigate band bending and carrier accumulation in the thinner p-waveguide, which is Fig.17 This is confirmed in the simulation results in . At high injection current, the power saturation effect is minimized and the efficiency is significantly improved due to the reduced optical losses.

[0082] To confirm the performance, a LOC laser diode based on Design Example 1 has been tested. Fig.18 Shown are measured voltages for several thin p-waveguide designs for a 976 nm laser diode fabricated with a stripe width of 200 μm (slow axis width). Up to 5.3% voltage reduction was observed in the thinnest p-waveguide design compared to the conventional LOC design. Fig.19 The measured fast-axis far field of the thinnest p-waveguide design is plotted, which shows not only comparable far-field divergence angles as in the conventional LOC design, but also single-mode operation in the vertical direction for all measured currents up to 25A.

[0083] Fig. 20Measured laser diode power and voltage as a function of drive current are shown for a conventional LOC design and for a laser diode design constructed according to Design 2a. Also shown are standard (i.e., see Fig.15 2a. The graph shows the voltage reduction achieved by Design 2a, based on a comparison of a standard device with a device implemented according to Design 2a. This is a result of making the p-waveguide thickness thinner.

[0084] Due to the voltage reduction, the efficiency associated with design 2a (see Fig.21 ) is higher. Specifically, Fig.21 The measured electro-optical power conversion efficiency as a function of drive current is shown for a conventional LOC design and a laser diode design constructed according to Design 2a. The electro-optical power conversion efficiency is defined as the output power divided by the product of voltage and current. Optical power is typically measured using a thermopile. Conventional LOC devices exhibit lower efficiency at currents of approximately 6.5A and higher.

[0085] Finally, if Fig. 22 As shown in the fast-axis far field, the device still maintains single-mode operation despite the significant thinning of the p-waveguide. Fig. 22 The measured fast-axis far field of a laser diode constructed according to Design 2a using the thinnest p-waveguide design is plotted, and single-mode operation in the perpendicular direction is also shown.

[0086] in conclusion

[0087] The skilled person will appreciate that many changes may be made to the details of the above-described embodiments without departing from the basic principles of the invention. Fig.23 Shown Figure 3 and Fig.12 In particular, graph 2300 shows a double n-cladding (similar to Figure 3 shown) and dual waveguides (similar to Fig.12 The combination of these features provides further model identification attributable to dual n-cladding filtering and distinction of limiting factors attributable to dual waveguides, as previously explained in connection with design examples 1a and 3, respectively. The scope of the present invention should therefore be determined solely by the appended claims.

Claims

1. A large optical cavity LOC laser diode (200) configured to support lasing of a fundamental mode (356) and to suppress lasing of first and higher order modes (370) along a fast axis of a beam emitted by the laser diode, the laser diode comprising: An optical cavity defined by a p-side (214, 254, 250, 326, 338) of a laser diode, an n-side (230, 220, 216, 310, 320) of the laser diode, and an active region (210, 340, 344) including a quantum well (344) located between the p-side and the n-side, the n-side including an n-waveguide layer (320) forming at least a portion of a multimode waveguide (350) having the quantum well (344) offset toward the p-side; and Adjacent claddings (316, 312, 716, 712) on one or both of the p-side and the n-side, the adjacent claddings comprising an outer cladding (312, 712) and an inner cladding (316, 716), the inner cladding being located between and adjacent to the multimode waveguide (350) and the outer cladding, the outer cladding and the inner cladding having first and second refractive indices, respectively, the first refractive index of the outer cladding being greater than the second refractive index of the inner cladding and greater than the effective refractive index of the first-order mode (370) so as to couple it out of the multimode waveguide. 2 . The LOC laser diode of claim 1 , wherein the n-side includes the adjacent cladding layer. 3 . The LOC laser diode of claim 1 , wherein the p-side includes the adjacent cladding layer.

4. The LOC laser diode of claim 1, wherein the p-side includes a p-cladding layer adjacent to the active region such that the n-waveguide includes substantially the entire portion of the multimode waveguide.

5. The LOC laser diode of any one of claims 1-4, further comprising a graded-index refractive index profile adjacent to the active region.

6. The LOC laser diode of any one of claims 1-4, further comprising a step-index refractive index profile adjacent to the active region.

7. The LOC laser diode of claim 4, further comprising an asymmetric graded-index refractive index profile adjacent to the active region.

8. The LOC laser diode of claim 4, further comprising an asymmetric step-index refractive index profile adjacent to the active region.

9. A large optical cavity (LOC) laser diode (200) configured to suppress lasing of first and higher order modes (1270, 1370) along a fast axis of a light beam emitted by the laser diode, the laser diode comprising: An optical cavity defined by a p-side (214, 254, 250, 1266, 1366) of a laser diode, an n-side (230, 220, 216, 1260, 1262, 1360, 1354, 1362) of a laser diode, and an active region (210) including a quantum well located between the p-side and the n-side; a p-waveguide layer, the p-waveguide forming at least a portion of a multimode dual waveguide (1250, 1350); said quantum well being offset toward said p-side; and a set of n-waveguide layers (1260, 1262, 1360, 1362) on the n-side, forming at least a portion of the multimode dual waveguide, the set of n-waveguide layers comprising an outer n-waveguide layer (1260, 1360) and an inner n-waveguide layer (1262, 1362), wherein optical gain applied to the first and higher order modes (1270, 1370) is reduced relative to optical gain applied to a fundamental mode.

10. The LOC laser diode of claim 9, wherein the inner n-waveguide layer is located between and adjacent to the active region and the outer n-waveguide layer, the outer n-waveguide layer and the inner n-waveguide layer having a first refractive index and a second refractive index, respectively, the first refractive index of the outer n-waveguide layer being less than the second refractive index of the inner n-waveguide layer.

11. The LOC laser diode according to claim 9 or 10, wherein the multimode dual waveguide has an asymmetric refractive index.

12. The LOC laser diode of claim 9 or 10, further comprising adjacent n-cladding layers, the adjacent n-cladding layers comprising outer and inner cladding layers, the inner cladding layer being located between and adjacent to the outer n-waveguide layer and the outer cladding layer, the outer and inner cladding layers having first and second refractive indices, respectively, the first refractive index of the outer cladding being greater than the second refractive index of the inner cladding layer and greater than the effective refractive index of the first-order mode to couple it out of at least the inner n-waveguide layer.

13. The LOC laser diode of claim 9, wherein the set of n waveguide layers forms at least a portion of a symmetric waveguide having matched refractive indices and separated by cladding layers having a refractive index less than that of the set of n waveguide layers.

14. The LOC laser diode of claim 13, wherein the cladding is offset to the n-side such that a first portion of the multimode dual waveguide is wider than a second portion of the multimode dual waveguide, the first portion including the inner n-waveguide and p-waveguide layers, and the second portion including the outer n-waveguide layer.

15. A diode laser assembly comprising the LOC laser diode according to claim 1 or 9.

Images