Partial Amplification of Octave-Spanning Supercontinuum in the Spectral Region of 1.5–2.2 μm

Round 1

Reviewer 1 Report

The manuscript is well written. The objective and the results are clearly presented. I strongly advice to add numerical study to this work. Author can also include a discussion based on comparing their study with high power supercontinuum generation with graded-index multimode fibers in the presented wavelength range. For some of the instance following papers are relevant for comparison,

• Teğin, Uğur, and Bülend Ortaç. "Cascaded Raman scattering based high power octave-spanning supercontinuum generation in graded-index multimode fibers." Scientific reports 8.1 (2018): 1-7.
• Krupa, Katarzyna, et al. "Spatiotemporal characterization of supercontinuum extending from the visible to the mid-infrared in a multimode graded-index optical fiber." Optics Letters 41.24 (2016): 5785-5788.

Author Response

We thank you for your feedback on our manuscript.

At the moment, we are trying to carry out numerical simulation of radiation propagation in DDF, and in the future we will be able to evaluate it also with partial amplification of the SC. However, that will be the next step in our investigations. Now  we can only present experimental data.

If compared with SC generators with graded-index multimode fibers, then when using MMF fibers, in general, an optical spectrum is obtained from 520 to 2800 nm [1*, 2*], and the output power can reach up to 1 W [3*]. The main difference of the system presented by us is the simplicity of design due to the use of only quartz fibers. That is, our generator is all-fiber without the use of complex and expensive elements, such as bulk optics, mode field adapters, etc. After optimizing all the components of the generator, it is also possible to increase the energy of the pulsed radiation.

The comparison has been added to table 2 (see the new manuscript version).

1* Eslami, Z., Salmela, L., Filipkowski, A. et al. Two octave supercontinuum generation in a non-silica graded-index multimode fiber. Nat Commun 13, 2126 (2022). https://doi.org/10.1038/s41467-022-29776-6

2* A. Niang, T. Mansuryan, K. Krupa, A. Tonello, M. Fabert, P. Leproux, D. Modotto, O. N. Egorova, A. E. Levchenko, D. S. Lipatov, S. L. Semjonov, G. Millot, V. Couderc, and S. Wabnitz, "Spatial beam self-cleaning and supercontinuum generation with Yb-doped multimode graded-index fiber taper based on accelerating self-imaging and dissipative landscape," Opt. Express 27, 24018-24028 (2019)

3* Krupa, Katarzyna, et al. "Spatiotemporal characterization of supercontinuum extending from the visible to the mid-infrared in a multimode graded-index optical fiber." Optics Letters 41.24 (2016): 5785-5788.

Reviewer 2 Report

Comments for author File: Comments.pdf

Author Response

We thank you for your feedback on our manuscript. It helps us to improve our manuscript quality. Below we provide a point by point reply to your comments.

A: In the initial literature review, we focused on laser systems only for the amplification of already generated SC. Therefore, other types of SC generators were not considered in this case. For improvement, Table 2 was added with an extended comparison of supercontinuum sources.

Line 25: The sentence, ‘The effect on biological tissues’ is not correct. To study or explore should be used. 

A: We thank you for your comment. This sentence has been corrected

Line 36: The manuscript is about 1.5 and 2 um, hence description of 2.5 to 2.8 um spectral region is irrelevant. 

A: This fact is mentioned as a disadvantage of silica-based systems.

Line 59 to 65: This paragraph is not relevant in the introduction. Even before describing the spectral feature of the amplifier, authors address the advantages and applications. This paragraph should be in the results and discussion section. How was the Dispersion decreasing fiber (DDF) made? Splice loss between the different fiber amplifiers has not been mentioned.

A: Thank you for the comment. The DDF fiber used in this work was produced in the course of joint research of GPI RAS and ICHPS RAS  using the classical technique of taper-type fiber drawing while maintaining a smooth change in the core diameter along the length of the entire sample (here from 6 to 9 um) [1*,2*]. The DDF fiber was doped with germanium oxide. D varies from 0 to 11 ps/km*nm at 1.5 um along the fiber. Splice losses were estimated by the equipment itself and the value did not exceed 0.01 dB.

 

1*V. A. Bogatyrev, M. M. Bubnov, E. M. Dianov, A. S. Kurkov, P. V. Mamyshev, A. M. Prokhorov, A. S. Rumyantsev, V. A. Semenov, S. L. Semenov, A. A. Sysoliatin, S. V. Chernikov, A. N. Guryanov, G. G. Devyatykh, and S. I. Miroshnichenko, “A single-mode fiber with chromatic dispersion varying along the length,” J. Lightwave Technol., vol. 9, pp. 561–566, 1991. doi:10.1109/50.79530

2*Mostofi, A., Hatami-Hanza, H., & Chu, P. L. (1997). Optimum dispersion profile for compression of fundamental solitons in dispersion decreasing fibers. IEEE Journal of Quantum Electronics, 33(4), 620–628. doi:10.1109/3.563391 

 Line 114: Optical spectra were controlled by listed optical spectrum analyzers. What is this means? Spectra can be measured or recorded with the aid of spectrum analyzers, not controlled. 

A: It means that we are monitoring the spectrum of the reference oscillator emission, and we also monitor the spectral change of the SC and its gain.

Line 134: where the pulse envelope duration was measured? At Full width half maximum or 1/e2 What is the reason for choosing active fiber lengths? What are the effects of master oscillator spectral characteristics on SC generation?

A: The envelope measurement was performed by FWHM. The active fiber length of the amplifiers was selected experimentally to obtain the most effective radiation gain. Experiments on the effect of the MO spectral characteristics changing on the SC spectra were carried out. 

fig.1. Optical spectra of MO (a) and SC (b) as a function of time (without control of MO).

fig.2. Optical spectra of MO (a) and SC (b) as a function of time (control of MO by changing the diode pumping current).

Figure 1 shows the spectral change of SC as a function of changes in MO for 10 minutes. Most of the changes occurred in the spectral regions of 1.3-1.4 μm, 1.5-1.8 μm. Here we can see the predominant influence of the Stokes component on the SC. If we controlled the MO by varying the pump current LD1 (fig.2), the changes were insignificant and the intensity changes did not exceed 2-3 dB.

Line 132: Evolution of SC generation is not described in detail, apart from mentioning it is due to stimulated Raman scattering (SRS) in combination with MI and solitons fission. 

A: For a more detailed description of the nonlinear effects prevailing in our case, a separate thorough mathematical simulation is necessary. At the moment, such a task is for the future. For the time being, we are guided by the already published materials on SC generation.

Line 133: While in sentence says the octave width is -30 dB, figure 2 is not on a similar scale to compare. 

A: Fig.3. shows the spectrum of broadband SC with a dB scale. As can be seen, at -30 dB the spectrum width extends from 905 to 2309 nm, which is an octave. However, in the manuscript, the spectrum in logarithmic scale with relative intensity is presented, for uniformity of all presented spectra. -30 dB in our case equals 0.001 level.

fig.3. Optical spectra of SC

Line 140: very hard to understand the meaning 

A: Two spectrum analyzers were used to obtain an output spectrum in a wide range and with good resolution (0.5 nm). Suggestion modified.

Line 147: what is the coupling and splice efficiency here? What is the power at the end of EDFA before amplification?

A: The splice losses were estimated by the equipment itself and the value did not exceed 0.01 dB. Passed radiation at the output of the amplifier was 18 mW. The input power was 340 mW.

Line 174: lacks the scientific quality of justification 

A: The sentence is changed (line 173-175).

Line 180: figure 4: what is the reason for significant spectral modulation? 

A: Significant changes in fine spectral structure (significant ruggedness of spectrum opposite to previous one) can be explained by another dispersion parameter of active fiber. 

Line 188: what is the coupling and splice efficiency here? What is the power at the end of TDFA before amplification? 

A: The splice losses were estimated by the equipment itself and the value did not exceed 0.01 dB. Passed radiation at the output of the amplifier is less than 1 mW. 

LINE 209: What is the reason for the lower input signal? In both EDFA and TDFA, pulse peak power parameters were not discussed. How is the comparison being made here in HDFA? 

A: We can only give an approximate estimate of the peak power for all amplifiers, since we only know the duration of the envelopes. The output power from the system without amplification was 4 mW. The low signal was due to the fact that we significantly reduced it for analysis (in this case, the average output power is several times higher) without the risk to damage the equipment.

Line 218: according to my understanding, HNLF is not used in this work? Why does it say HNLF? 

A:  Thanks for the comment, corrected to DDF.

Line 222: Table 1: Spectral bandwidths: in the text corresponding to SC, 1250 nm @-50dB was mentioned. While in the table says1260 nm at -20dB here? Similarly to the case of all three fiber amplifiers, spectral bandwidth after amplification is not described in the text

A: There was an error in the text. It meant 1260 nm at -20 dB. The spectral width obtained after amplification is not described, because the table contains all the information.

Author Response File: Author Response.pdf

Reviewer 3 Report

This manuscript concerns partial amplification of a supercontinuum spectrum using three different types of rare-earth doped fiber. The experiments and results are quite simple. The only novelty and impact of this manuscript lies in the use of these rare-earth doped fibers to partially amplify the supercontinuum, so as to increase spectral power density beyond the pump laser wavelength. However, the spectral power density of the three cases ranges from 1 mW/nm to 6.6 mW/nm. Compared with the state-of-art fibers, the spectral power density is not improved. When compared with commercially available system such as NKT Photonics (https://www.nktphotonics.com/products/supercontinuum-white-light-lasers/superk-fianium/) , especially for the 1.5 um region, where the power spectral density is much lower reported here. Not only extra pump lasers are needed, but also the bandwidth and power spectral density are potentially sacrificed.

So my question is, what are the advantages of using partial amplification here? Can’t an engineered highly nonlinear fiber provide better results? 1 um pump is widely used to generate broad bandwidth, but for the wavelength range that the authors care about, shouldn’t pumping at 1.5 um give a better result?

Apart from the main concerns, here are my detailed suggestions for this manuscript:

1. The authors mentioned “It is worth pointing out that home-made dispersion decreasing fiber”, so what is the home-made dispersion decreasing fiber? What is the dispersion of the fiber? The authors should add the dispersion plot if it is worth pointing out.
2. Since power spectral density is important, the authors should add figures of the power spectral density.
3. What is the power of LD1 laser?
4. The corresponding author was wrongly marked.
5. “FA” aberration appears before the mention of “fiber amplifier”.
6. Some parts are really hard to read, the authors should extensive edit the manuscript.

Author Response

We thank you for your feedback on our manuscript. It helps us to improve our manuscript quality.

We agree that the initial SC has lower than commercial ones spectral power density, but our approach allowed us to improve this parameter in the chosen spectral ranges. 

The fibers used provide the simplicity of laser system design without the use of various additional components (e.g., mode field adapters), and we do not use here the bulk optics, which is necessary for such types of NHLF as some types of ZBLANs, chalcogenides and PCFs. In the future, there will be experiments comparing the use of pumping in different spectral regions. This includes the use of pumping at 1.5 µm as well.

Below we provide a point by point reply to your comments.

1. The authors mentioned “It is worth pointing out that home-made dispersion decreasing fiber”, so what is the home-made dispersion decreasing fiber? What is the dispersion of the fiber? The authors should add the dispersion plot if it is worth pointing out.

A: The DDF fiber used in this work was produced in the course of joint research of GPI RAS and ICHPS RAS  using the classical technique of tepper-type fiber drawing while maintaining a smooth change in the core diameter along the length of the entire sample (here from 6 to 9 μm). The DDF fiber was doped with germanium oxide. D = 0-11 ps/km*nm at 1.5 um.

1. Since power spectral density is important, the authors should add figures of the power spectral density.

A: Thank you for your comments, this comment has been noted in the manuscript.

1. What is the power of LD1 laser?

A: The diode output power used did not exceed 4 W to realize the required laser mode of operation. The maximum output power is 8 W.

1. The corresponding author was wrongly marked.

A: Thank you for your comment. The corresponding author is now correctly marked.

1. “FA” aberration appears before the mention of “fiber amplifier”.

A: Thank you for your comment. The manuscript has been corrected.

1. Some parts are really hard to read, the authors should extensive edit the manuscript.

A: The manuscript was revised and the changed parts are highlighted.

Reviewer 4 Report

This manuscript shows the experimental results about the octave-spanning supercontinuum generation from three different rare-earth-doped fiber amplifiers. This article is interested and well-organized. So, i would recommend this article for publication in this journal with the following some minor revision. 

1. In figure 1, is the used Yb-GTWave fibers homemade or commercial?
2. In line 111 on page 2, what is the "PL" meaning?
3. Did the authors measure the stability of the output  (i.e., long-term stability or power stability)? I wonder the stability of the obtained SC.  
4. Can the authors measure the output spatial distribution using a beam profiling camera?

Author Response

We thank you for your feedback on our manuscript. It helps us to improve our manuscript quality. Below we provide a point by point reply to your comments.

• In figure 1, is the used Yb-GTWave fibers homemade or commercial?

A: The Yb-GTWave used are homemade samples.

• In line 111 on page 2, what is the "PL" meaning?

A: This abbreviation meant “pump laser”. A correction has been made to the manuscript.

• Did the authors measure the stability of the output  (i.e., long-term stability or power stability)? I wonder the stability of the obtained SC.  

A: The stability of the obtained SC depending on the laser operation was assessed. Also, the power changes were monitored for 15-20 minutes, which varied between 330-350 mW. Longer measurements were not carried out.

• Can the authors measure the output spatial distribution using a beam profiling camera?

A: Unfortunately, we do not have this device, so we cannot estimate the output spatial distribution.

Author Response File: Author Response.pdf

Reviewer 5 Report

This work presents a supercontinuum light source analysis based on different rare-earth-doped fiber lasers. The authors analyze the difference between different active mediums in a fiber laser cavity to generate several wideband spectrums. The fiber laser cavity is interesting, and the wideband spectrums can be used for different applications. However, there are some points that the authors need to attend to before acceptance in PHOTONICS.

1. The authors need to present the evolution of the wideband as the pump power is increased for each rare-earth-doped fiber studied. Please include this information and its discussion. This information can be included in the final spectrum achieved for each rare doped fiber analyzed or by a graph representing the input power vs. wideband.
2. Although the authors demonstrate an SC generation, the spectrum is not flat; this is a desired characteristic of the SC light sources. The authors need to discuss this issue. It is a surprise that the authors do not mention this issue in the manuscript.
3. The authors need to mention specific applications of the SC spectrums achieved. As the previous comment states, the poor flatness limit the possible applications. 
4. The authors need to highlight their contribution to the literature. A comparative table in terms of wideband and wavelength rage is strongly suggested. Please consider prior works to elaborate the material. 

Author Response

We thank you for your feedback on our manuscript. It helps us to improve our manuscript quality. Below we provide a point by point reply to your comments.

1. The authors need to present the evolution of the wideband as the pump power is increased for each rare-earth-doped fiber studied. Please include this information and its discussion. This information can be included in the final spectrum achieved for each rare doped fiber analyzed or by a graph representing the input power vs. wideband.

A: Two spectra were demonstrated for each amplification variant. One at minimum pumping power, the other shows the final conversion.  Figure 1 (see attached file) shows the dependence of the amplifiers' pumping power on the width of the resulting spectrum at a signal level of -20 dB. But we think that these figures will not improve the manuscript.

2. Although the authors demonstrate an SC generation, the spectrum is not flat; this is a desired characteristic of the SC light sources. The authors need to discuss this issue. It is a surprise that the authors do not mention this issue in the manuscript.

A: Not all applications require a flat spectrum in a wide range. Here we considered applications specifically for the amplified parts of SC, namely applications in the 1.5-2.1 μm spectral region. Therefore, the asymmetry appearing in the SC spectrum, allows us to obtain the greater energy in the necessary part of the optical spectrum.  As an example, such a scheme can serve for the development of a mobile endoscope, which would be used at the place of medical care. A separate issue is also the study of different tissues in the human body. For example, there are absorption lines of lipids in the spectral region of 2.3 μm, and our data are suitable for this type of research. 

3. The authors need to mention specific applications of the SC spectrums achieved. As the previous comment states, the poor flatness limit the possible applications. 

A: Possible applications of the received radiation were considered in lines 59-65. Since the SC generator has not been optimized, we have not yet used it in practice. However, under laboratory conditions we carried out studies on carbon nanotubes, as well as illuminated fiber Bragg gratings in the spectral range of 1.9-2.1 μm.

4. The authors need to highlight their contribution to the literature. A comparative table in terms of wideband and wavelength rage is strongly suggested. Please consider prior works to elaborate the material. 

A: We have taken into account your comment. New table was added to the manuscript.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

-

Author Response

Thank you for your review!

We improved our text.

All corrections are highlighted in yellow.

Author Response File: Author Response.pdf

Reviewer 3 Report

The authors didn’t fully address my concerns. Regardless of the impact and novelty, the manuscript still lacks the info I requested. 

 

As I mentioned in the previous review, since power spectral density is important, the authors should add figures of the power spectral density. A single number can’t represent  the whole spectrum, so adding a single number doesn’t really make any sense. 

 

In addition, DDF is an important element in the experiment, so its broadband dispersion characteristic is needed to evaluate and further simulate the SC. A single dispersion number at 1550 nm is not enough, a broadband dispersion plot is needed.

 

I would only support publishing if these two aspects are satisfied.

Author Response

Thank you for your rewiev!

Figure 6 in the manuscript, the optical spectrum has been replaced.

The DDF dispersion plot is entered in Figure 2.

Author Response File: Author Response.pdf

Reviewer 5 Report

In this new version, most of the comments were attended. However, the first remark is not addressed. The authors need to show the spectrum response considering the pump power. 

 

A: Two spectra were demonstrated for each amplification variant. One at minimum pumping power, the other shows the final conversion. Figure 1 shows the dependence of the amplifiers' pumping power on the width of the resulting spectrum at a signal level of -20 dB.

Reply: The graphs should be “pump power vs. bandwidth”; the opposite way does not provide clear information. Please fix these graphs and include them in the manuscript. 

Author Response

Thank you for your review!

Mentioned graphs have been corrected and added to the manuscript (Fig. 3(e), Fig. 4(e), Fig. 5(f)).

Author Response File: Author Response.pdf