A Directly Modulated Laterally Coupled Distributed Feedback Laser Array Based on SiO₂ Planarization Process

Round 1

Reviewer 1 Report

Yes, it is true that electrode capasitance can limit the modulation bandwidth.  However, there are quite many parameters that also affect the bandwidth.  For example junction capasitance also contributes to the total capasitance.   Figure 3a is quite blurry. A sharper picture is required.    Yes, the modulation bandwidth is wide, but also the response is not very flat.  Therefore, the large signal modulation may generate bit pattern dependent behaviour. Hence, in large signal modulation the eye diagram can be actually closed much earlier than the small singal modulation predicts.   On line 114 "Such excellent output power performance is attributed to combination  of good profile of lateral gratings" can you explain why.   On line 117 "The value of κ is estimated to be 26.7 cm-1" how the value is estimated?   On line 72 "Note that large optical cavity is helpful to improve ?", is this statement valid?  Usually large optical cavity in the vertical direction does not actually improve QW confinement factor.  Large optical cavity increases the optical confinement in the waveguides. Narrow waveguides usually leak more power to the outside the waveguides. This improves the grating coupling coefficient of LC-DFB lasers.       On line 99, how is the cavity length chosen?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper described high bandwidth operation of high duty cycle 3rd order grating LC-DFB lasers emitting around 1.3µm band. The grating is defined with electron beam lithography and the fabrication details are given in enough detail. The presentation is quite clear, but needs some refinement. Most striking points are listed below in as more detailed comments.

Grammar of the whole paper should be rechecked, especially there are mistakes in singular/plural forms of nouns/verbs in several places. For example,
"Low-cost and high performance laser source, such as distributed feedback (DFB) laser source, is urgently needed."
should be corrected to either
"Low-cost and high performance laser sources, such as distributed feedback (DFB) laser sources, are urgently needed."
or
"A low-cost and high performance laser source, such as a distributed feedback (DFB) laser source, is urgently needed."

Please add citations to any modulation results from LC-DFB tests, see for instance https://doi.org/10.1007/s11082-016-0871-9

"A P-type electrode pad with a diameter
58 of 100 μm introduces a parasitic capacitance C of 0.18 pF." How is the p-pad parasitic capacitance calculated?

"The relaxation oscillation frequency is given by" Citation is needed in the relaxation frequency formula.

"Note that large optical cavity is helpful to improve Gamma" Usually a large optical cavity is used to mean the waveguide in the epitaxial direction. Such a waveguide widens the optical mode in the epitaxial direction, consequently decreasing Gamma. Please clarify what you mean here.

"while a narrow waveguide is needed for LC-DFB to increase the grating coupling coefficient kappa" Does narrow waveguide refer to the width of the ridge or, in comparison to the beginning of the sentence, to the epitaxial dimension of the waveguide? Also, why does the narrow ridge increase kappa?

" The relationship kappa and Gamma versus the ridge width are calculated by finite element analysis (FEA) method as shown in Figure 1(b)." Please give a reference and note if the method has some limitations. FEA is probably used to calculate the mode profiles, but it's not enough on its own to evaluate these quantities.

Figure 3: The SEM figures have suffered from compression, please use better versions if possible.

Figure 4a: Are there any mode hops over the used bias range? It would be good to see the peak wavelength as a function of bias current alongside the power level, for example.

"The value of kappa is estimated to be 26.7 cm-1" How is the measured kappa estimated?

"The 3-dBe bandwidths of the small-signal modulation response shown in Figure 5 exceed 14 GHz for all four channels." There is quite a strong carrier resonance peak in the modulation response around 10 GHz. Please comment if this would be a problem in actual data transfer applications (=eye diagram / bit-error rate measurements).

"The D factors" D factor should be defined and a citation used.

"The values of four-channel differential gain are obtained by" Citation for differential gain formula needed.

Author Response

Please see the attachment.

Author Response File: Author Response.docx