Highly Efficient Nanosecond 1.7 μm Fiber Gas Raman Laser by H2-Filled Hollow-Core Photonic Crystal Fibers
Round 1
Reviewer 1 Report
The manuscript report on demonstration of an ns-pulsed fiber gas Raman laser emitting at 1.7 μm. Attractive features of the laser are 1) all-fiber configuration, 2) slope efficiency above 80%, 3) tunable wavelength, and 3) average power as high as 3 W. The manuscript is clearly written and the reference list is adequate. I would suggest accepting the manuscript for publication provided the following comments will be addressed:
1) Although general ideas flow is quite clear, the text should be carefully revised in terms of English. Some examples that require correction are:
page 2, line 60. "because" should be deleted
page 2, line 63. "can up to 1.5 W" should be replaced by "can be up to 1.5 W"
page 2, line 67. "has been demonstrated to operating" should be replaced by "has been demonstrated to operate"
page 2, line 75. the word "fibers" should be deleted
and many other minor mistakes. Please, revise the text once more.
2) Mode field diameters of SMF-28e and HC-1550-02 fibers are very close. The question is why the splice loss is as high as 1.4 dB.
3) H2-molecule is known to have two forms - ortho-hydrogen (rotational Raman shift of 587 1/cm) and para-hydrogen (rotational Raman shift of 354 1/cm) Please comment on why the pump wave was not scattered on para-H2 rotations? When first Stokes wave at 1.7 μm becomes strong enough to serve as a pump for the next SRS cascade, why only rotational SRS in para-H2 was observed? (Note, both 354 and 587 1/cm shifted waves fall into high loss region of the HCF).
4) The 1st Stokes power reaches 3 W. The 2nd Stokes power is 20 dB lower according to Fig.2b, so its power should be around 30 mW, which is measurable power level. How the Authors explain that no power at 1.8 μm was detected by a powermeter.
5) regarding the theoretical model: please, provide the values that were used for the Raman gain in 1st and 2nd cascades. How have you calculated those values? Also, the model requires some input "seed" power at the wavelength of 1st and 2nd Stokes; please comment on how you choose these input boundary conditions.
6) when repetition rate is below 1.3 MHz, the 1st Stokes wave loses the power (at high pump levels). What is the reason for such behavior? Where the power flows to?
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments on “Highly Efficient Nanosecond 1.7 μm Fiber Gas Raman Laser by H2-filled Hollow-core Photonic Crystal Fibers” by HAo Li et al.
The paper describes the construction and performances of a laser device aimed to generate pulsed radiation at 1.7 microns by means of Stimulated Raman Scattering in a H2 filled photonic crystal fiber.
Overall the paper is well written and the results are interesting for the development of similar devices. Nonetheless, there are some aspect that should be improved or reported with greater details, in particular:
1) As the Authors present an extensive characterization of the behavior of the device for changing the pump pulse repetition rate, they should provide some information regarding the pump pulse length and shape at the input of the hollow core fiber. Actually, the Authors state that “the pulse width is 15 ns” (line 105) but it is not clear if this is the pulse length just after the modulator or at the end of the amplification, and if this value changes at different output power levels at a given pulse repetition frequency.
2) Regarding the results presented in Figure 4 (b) and (c), the Authors should add at least one graph where the output pulse energy (or the conversion efficiency) is reported as a function of the input pulse energy (rather than as a function of the input average power), for the different pump repetition rates used in the experiment. This would show more clearly the behavior of the conversion process as a function of the pump pulse energy (and thus as a function of the peak power, as long as the pulse duration is constant, see the point above). In the current representation of the data this important information is provided only indirectly because the data are shown as a function of the pump average power.
3) In the discussion of the model of Eq. 1, the Authors should specify which values of the relevant parameters (i.e. the Raman gain and the values of the losses at the various wavelengths) they have used. Are these values determined independently or they have been adjusted in some way to fit with the experimental data? Please specify and discuss.
4) Regarding the conversion on the 2nd Stokes order: the Authors state that “Owing to the high fiber loss, the second-order Raman line is too weak to be measure by power meter” (lines 146-147). But looking at the spectrum of Figure 2b (measured at the maximum pump power and at 1 MHz of rep rate), it appears that the output at 1800 is about 22 dB lower (i.e. a factor of ~13) than the main output at 1700 nm. By looking at the data of Figure 4b, the output power on the first Stokes order in these conditions is about 2.6 W, so the output at 1800 nm should correspond to approximately 200 mW. This is not negligible with respect the output on the 1st stokes order, and should be easily measurable with a standard power meter, with a proper filtering of the residual pump and 1st Stokes radiation. Indeed the data of Figure 4 show power levels which are even less than 200 mW. The Authors should clarify if they have actually tried to measure the 2nd stokes output, and with which results. If the 2nd stokes output is actually undetectable with the power meter, they should investigate the reason of the discrepancy with the spectrum of Figure 2 b.
5) In figure 1 (a) the mirror M is probably shown in a wrong position, please fix.
With these modifications the paper can be published without further revisions.
Author Response
Please see the attachment
Author Response File: Author Response.docx
