An Integrated Germanium-Based THz Impulse Radiator with an Optical Waveguide Coupled Photoconductive Switch in Silicon
Round 1
Reviewer 1 Report
The authors report a novel method to generate THz signal pulse by utilizing Germanium-medium integrated with silicon waveguide. The highly doped Ge thin film is applied to effectively suppress photon-generated carriers to obtain short THz pulses. It is a significant improvement to generate ultra-fast THz pulse single by using photo-conductive switch. And the author provides solid measurement result to support their claims in manuscript. Here I have a few comments for authors to answer. Please find them as below:
1. The authors claim that “we present a Germanium (Ge)-based THz impulse radiator in silicon that resolves 70 the aforementioned limitations”. Is this approach totally novel in aspect of fundamental mechanism to generation THz pulse single? What is the difference between the authors’ approach and the one used in reference 17 in term of fundamental mechanism to generation THz pulse single?
2. The authors utilize an on-chip integrated waveguide to couple incident 1550nm excitation laser pulse. Since the incident light is a ultra-fast laser pulse, does dispersion affect overall performance, such as conversion efficiency, when the pulse propagates in the silicon waveguide?
3. The authors measured the photon-generated THz single in both time and frequency domain and claim that “the measured results confirmed this theoretical prediction” in line 173. Could the author provide the theoretical model and result that can be compared with its measured results? Meanwhile, the authors claim that the conversion efficiency is 3.6 × 10-5. How does the author get this number? In the optical measurement setup, optical coupling loss exists in free-space to chip coupling for excitation light as well as chip-to-THz detector for generated THz pulse. Is the coupling loss excluded in conversion efficiency calculation?
4. In Figure 8 (c), what is the measured standard deviation?
Author Response
Thank you for the positive comments on our manuscript. We addressed all your comments and made revisions in the manuscript as much as we can. All the revisions in the manuscript are marked in blue for your convenience. Please refer to the attached document for our detailed response.Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors report on the design, fabrication and characterization of an integrated THz pulse radiator based on a Ge-on-SOI chip. Overall, their results seem convincing to me and I think they might be of interest for the community; moreover the paper is well written and easy to read. I therefore recommend the publication of the manuscript, provided the authors answer a few comments:
1) In line 129 the authors claim that that the Ge film was design to optimize both absorption and conversion efficiencies. However, figure 4 only shows simulations for absorption optimization. Can the authors add a figure or describe more in detail the conversion efficiency optimization calculations?
2) In line 160 the authors state that they adjusted their sample on a rotational stage so as to maximize the coupling efficiency to the grating. Actually what efficiency can they achieve? Actually this quantity also depends on the NA of the lens focusing the laser on the chip. What is the NA of the lens?
3) The authors claim that the THz pulse generation process is due to photoconductive processes. However, papers such as Phys Rev Lett 68, 1 (1992) and Nano Energy 46, 128 (2018) also report on THz pulse generation through optical rectification induced by surface third-order nonlinearities, which in some cases can be efficient. Since Ge exhibits third-order optical nonlinearities I would expect this effect to take place. Can the authors comment whether they observed such phenomenon and whether it is relevant in their experiments?
Finally, I found a few typos while reading the manuscript. In any case I suggest the authors to carefully check if there are others that I missed.
Line 21: "THz impulses features" as the subject is plural, the verb needs no "s".
Line 24 "wirless" I guess the authors mean wireless
Line 109 "scheme. In this scheme" a bit repetitive
Line 146: "Metal vias" I guess the authors mean V bias
Author Response
Thank you for the positive comments on our manuscript. We addressed all your comments and made revisions in the manuscript as much as we can. All the revisions in the manuscript are marked in blue for your convenience. Please refer to the attached document for our detailed response.
Author Response File: Author Response.pdf
