Direct experimental verification of the predicted semiconductor-laser line forms and extension of the amplitude-fluctuation line-form model to VCSEL and microcavity lasers

A 2012 conference paper by M. G. Noppe derived a formula for the line form $L(\nu)$ of semiconductor-laser radiation in which spontaneous-emission amplitude fluctuations, entering through a field-dependent refractive index, add non-Lorentzian terms ($B_A\tau^2$, $C_A|\tau|^3$) to the usual phase-fluctuation broadening, and used it to simulate the measured natural linewidth $\Delta\nu$ of three Fabry–Perot lasers, including the well-known minimum in $\Delta\nu$ versus inverse output power. The paper left open three items: application of the model to other laser structures (distributed-feedback (DFB), vertical-cavity surface-emitting (VCSEL), and microcavity lasers) with determination of their parameters; direct measurement of the predicted line forms and comparison with the calculated curves; and model-based recommendations for minimizing $\Delta\nu$. Follow-up work by the same author completed the DFB extension (2014) and offered qualitative linewidth-reduction recommendations (2016), but the predicted line shapes have never been measured directly for any laser type, the VCSEL and microcavity cases have never been treated, and no independent group has tested, applied, or refuted the model. The surviving open core — experimental line-form verification plus VCSEL/microcavity extension — remains open.

Background

The natural linewidth of a laser is the width of its emission spectrum set by quantum fluctuations, originally estimated by the Schawlow–Townes formula, which predicts a linewidth decreasing monotonically as the inverse of the output power $P$ (Schawlow and Townes 1958). For semiconductor lasers, Henry's theory (Henry 1982) enlarged this by the linewidth enhancement factor $\alpha$, the ratio of carrier-induced refractive-index change to gain change, but still predicts a monotonic decrease of $\Delta\nu$ with increasing power. Experiments have long contradicted this monotonic picture: measurements on single-frequency (GaAl)As diode lasers showed a power-independent residual linewidth (Welford and Mooradian 1982); coherence measurements on gain- and index-guided lasers (in which the optical waveguide is formed by the gain distribution or by a built-in refractive-index step, respectively) found linewidth floors and upturns at high power (Elsässer, Göbel, and Kuhl 1983); and similar deviations were reported for separate-confinement-heterostructure (SCH) quantum-well distributed-feedback (DFB) lasers (Takano et al. 1989; a DFB laser uses a built-in corrugated grating rather than cleaved facets for wavelength-selective feedback). Linewidth data also exist for vertical-cavity surface-emitting lasers (VCSELs, which emit along the axis perpendicular to the chip surface) and for microcavity lasers (wavelength-scale resonators), but they do not show the same high-power deviations: Fordell and Lindberg 2007 measure a single VCSEL's linewidth in order to extract its linewidth enhancement factor $\alpha$, and the non-monotonic linewidth evolution that Bagheri et al. 2009 report near threshold — a local maximum a little above threshold, followed by narrowing at higher pump power — is attributed within the Henry framework to the linewidth enhancement factor and spontaneous-emission coupling, not to an amplitude-fluctuation upturn at high power. Mainstream partial explanations include occupation fluctuation noise, i.e. broadening by spontaneous fluctuations of the carrier population (Vahala and Yariv 1983) and 1/f carrier noise (Ohtsu and Kotajima 1984); none of these, however, was claimed to reproduce the full non-monotonic $\Delta\nu(1/P)$ curves, and a review position cited in this literature holds that the mechanisms of the natural linewidth remain in many aspects unclear.

Against this background, M. G. Noppe (building on an earlier derivation with A. P. Kovalevski, 2008, which had neglected a quadratic term in the phase-fluctuation correlation) derived a line-form formula for the radiation inside and outside the laser cavity: the spectrum is the Fourier transform of the field autocorrelation with exponent $A_p|\tau| + B_A\tau^2 + C_A|\tau|^3$, where $A_p = A_s^2/(4E^2\Delta t)$ comes from phase fluctuations of the spontaneous emission ($A_s$ the spontaneous fluctuation amplitude, $\Delta t$ the mean time between spontaneous emission events, $E$ the field amplitude) and $B_A = (\varphi A_s)^2/8$, $C_A = (\varphi A_s)^2/(12\Delta t)$ from amplitude fluctuations, the parameter $\varphi$ depending on the laser structure through a field-dependent refractive index $n = n_0 + n_1|E| + n_2E^2$ (an effective nonlinear photorefractivity in the sense of Partovi and Garmire 1991). The model's physical picture of the measured linewidth minimum is a competition: the phase term $A_p$ falls with power as $1/E^2$ while the amplitude term grows, so $\Delta\nu(1/P)$ bends upward at high power. In the 2012 NUSOD paper the model was applied to three Fabry–Perot lasers — an SCH laser (simulated minimum $\approx 1.72$ MHz at $1/P = 0.3\,\mathrm{mW}^{-1}$, compared with Takano et al. 1989) and the gain-guided V-groove and oxide stripe lasers (compared with Elsäßer et al. 1983) — and line forms were computed at three power values, but no direct spectral measurement of $L(\nu)$ was performed, and the DFB, VCSEL, and microcavity measurements cited as motivation were not simulated.

Problem Statement

Within the line-form model of Noppe (NUSOD 2012), in which the semiconductor-laser spectrum is the Fourier transform of an autocorrelation with phase-fluctuation and amplitude-fluctuation terms $A_p|\tau| + B_A\tau^2 + C_A|\tau|^3$ and a structure-dependent parameter $\varphi$ fixed by the field-dependent refractive index $n = n_0 + n_1|E| + n_2E^2$, the following parts of the source question remain open after the author's own 2014–2016 follow-ups:

1. Extension to VCSEL and microcavity lasers. Derive the structure-specific expressions for $\varphi$ and the line-form parameters for vertical-cavity surface-emitting and microcavity lasers (the DFB case having been treated in 2014), determine their numerical parameter values, and determine whether the model reproduces the published $\Delta\nu(1/P)$ curves for these structures — including Fordell and Lindberg's VCSEL data and Bagheri et al.'s near-threshold microcavity data — including any minimum or upturn in those curves.

2. Direct experimental verification of the line forms. Measure the laser line form $L(\nu)$ directly — e.g., by heterodyne or delayed self-heterodyne spectroscopy (in which the laser field is mixed with a delayed, frequency-shifted copy of itself so that the photocurrent power spectrum reproduces the laser line shape), with resolution and dynamic range sufficient to reveal the predicted non-Lorentzian structure — for operated lasers to which the model is applied, at output powers covering the regime of the predicted linewidth minimum, and compare the measured spectra quantitatively with the curves calculated from the model. This is the central untested prediction: the 2012–2016 works compare only computed linewidth values with published linewidth data, never the full spectral shape. Because no direct measurement of $L(\nu)$ exists for any laser type, the deliverable for this item is the prediction that such a measurement would test — the model's full line forms $L(\nu)$ for the structures of item 1 across the power range of the linewidth minimum, together with the measurement configuration (spectral resolution, dynamic range, and power range) needed to reveal the predicted non-Lorentzian structure — while the measurement itself remains an experimental objective.

3. Linewidth minimization. From the validated model, derive and state quantitatively which structural and material parameter choices minimize $\Delta\nu$, beyond the qualitative recommendations already published in 2016.

The common objective is to establish whether the amplitude-fluctuation line-form model is a quantitatively correct description of semiconductor-laser spectra across laser structures. Because the full line form has not been measured for any laser type, this claim is tested here through the model's predicted $\Delta\nu(1/P)$ curves, including the non-monotonic minimum and any upturn, against the published VCSEL and microcavity data. A negative answer is admissible and would be delivered the same way: a demonstration that the model cannot reproduce the measured line forms, or the VCSEL/microcavity linewidth data, resolves the common objective in the negative. Item 1 is therefore the resolving deliverable — a failure to reproduce the published VCSEL/microcavity linewidth data yields that negative answer — while item 2 states the line-form prediction that a future direct measurement would test and is consequently not a resolving condition in the absence of such measurements, and item 3 presupposes a model already validated by item 1 and is likewise not a resolving condition.

References

1. M. G. Noppe, 'On the line form and natural linewidth; Simulation and interpretation of experiments', 2012 12th International Conference on Numerical Simulation of Optoelectronic Devices (NUSOD), pp. 123–124 (2012). DOI: 10.1109/NUSOD.2012.6316548, https://doi.org/10.1109/NUSOD.2012.6316548
2. M. G. Noppe, 'On natural linewidth of Fabry-Perot lasers; simulation of experiments', 2014 12th International Conference on Actual Problems of Electronics Instrument Engineering (APEIE), pp. 468–471 (2014). DOI: 10.1109/APEIE.2014.7040723, https://doi.org/10.1109/APEIE.2014.7040723
3. M. G. Noppe, 'On natural linewidth of distributed feedback lasers; simulation of experiments', 2014 12th International Conference on Actual Problems of Electronics Instrument Engineering (APEIE), pp. 456–460 (2014). DOI: 10.1109/APEIE.2014.7040721, https://doi.org/10.1109/APEIE.2014.7040721
4. M. G. Noppe, 'The natural linewidth of Fabry–Perot semiconductor lasers; simulation of experiments', Laser Physics 24, 125006 (2014). DOI: 10.1088/1054-660X/24/12/125006, https://doi.org/10.1088/1054-660X/24/12/125006
5. Michael G. Noppe, 'On the nonlinear theory of Fabry–Perot semiconductor lasers', Laser Physics 26, 055004 (2016). DOI: 10.1088/1054-660X/26/5/055004, https://doi.org/10.1088/1054-660X/26/5/055004
6. Michael G. Noppe, 'On fundamentals of nonlinear theory for semiconductor lasers', 2016 13th International Scientific-Technical Conference on Actual Problems of Electronics Instrument Engineering (APEIE), pp. 382–384 (2016). DOI: 10.1109/APEIE.2016.7802304, https://doi.org/10.1109/APEIE.2016.7802304
7. Artyom P. Kovalevski and Michael G. Noppe, 'On new formulas for line form and natural linewidth in semiconductor lasers', 2008 9th International Conference on Actual Problems of Electronic Instrument Engineering (APEIE), pp. 83–88 (2008). DOI: 10.1109/APEIE.2008.4897064, https://doi.org/10.1109/APEIE.2008.4897064
8. A. L. Schawlow and C. H. Townes, 'Infrared and Optical Masers', Physical Review 112, 1940–1949 (1958). DOI: 10.1103/PhysRev.112.1940, https://doi.org/10.1103/PhysRev.112.1940
9. C. Henry, 'Theory of the linewidth of semiconductor lasers', IEEE Journal of Quantum Electronics 18, 259–264 (1982). DOI: 10.1109/JQE.1982.1071522, https://doi.org/10.1109/JQE.1982.1071522
10. D. Welford and A. Mooradian, 'Output power and temperature dependence of the linewidth of single-frequency cw (GaAl)As diode lasers', Applied Physics Letters 40, 865–867 (1982). DOI: 10.1063/1.92945, https://doi.org/10.1063/1.92945
11. W. Elsasser, E. Gobel, and J. Kuhl, 'Coherence properties of gain- and index-guided semiconductor lasers', IEEE Journal of Quantum Electronics 19, 981–985 (1983). DOI: 10.1109/JQE.1983.1071968, https://doi.org/10.1109/JQE.1983.1071968
12. Kerry Vahala and Amnon Yariv, 'Occupation fluctuation noise: A fundamental source of linewidth broadening in semiconductor lasers', Applied Physics Letters 43, 140–142 (1983). DOI: 10.1063/1.94260, https://doi.org/10.1063/1.94260
13. Motoichi Ohtsu and Shinichi Kotajima, 'Derivation of the Spectral Width of a 0.8 µm AlGaAs Laser Considering 1/f Noise', Japanese Journal of Applied Physics 23, 760–764 (1984). DOI: 10.1143/JJAP.23.760, https://doi.org/10.1143/JJAP.23.760
14. S. Takano, T. Sasaki, H. Yamada, M. Kitamura, and I. Mito, 'Sub-MHz spectral linewidth in 1.5 µm separate-confinement-heterostructure (SCH) quantum-well DFB LDs', Electronics Letters 25, 356–357 (1989). DOI: 10.1049/el:19890247, https://doi.org/10.1049/el:19890247
15. Thomas Fordell and Åsa Marie Lindberg, 'Experiments on the Linewidth-Enhancement Factor of a Vertical-Cavity Surface-Emitting Laser', IEEE Journal of Quantum Electronics 43, 6–15 (2007). DOI: 10.1109/JQE.2006.884583, https://doi.org/10.1109/JQE.2006.884583
16. Mahmood Bagheri, Min Hsiung Shih, Sang Jun Choi, John D. O'Brien, and P. Daniel Dapkus, 'Microcavity Laser Linewidth Close to Threshold', IEEE Journal of Quantum Electronics 45, 945–949 (2009). DOI: 10.1109/JQE.2009.2016850, https://doi.org/10.1109/JQE.2009.2016850
17. Afshin Partovi and Elsa M. Garmire, 'Band-edge photorefractivity in semiconductors: Theory and experiment', Journal of Applied Physics 69, 6885–6898 (1991). DOI: 10.1063/1.347678, https://doi.org/10.1063/1.347678