Numerical Analysis of a Dual-Wavelength-Clad-Pumped 3.5 μm Erbium-Doped Fluoride Fiber Laser
Round 1
Reviewer 1 Report
In this paper, the authors discuss the core pumping setup of the 1976 nm pump and its impact on stability using a 3.5 m erbium-doped fluoride fibre. They also compare the results with what is referred to as a more reliable and efficient dual-wavelength clad-pumped scheme that is analyzed numerically. The authors have described and compared both the traditional method of setting up a dual-wavelength pump as well as the method that the authors claim to be the "new" method, which involves optimizing the parameters. The authors propose a clad-pumped laser diode as an alternative to a core pumping setup, and they carry out various simulations in order to examine the properties of this new device. Although a lot of effort was put in, the outcomes do not convince me too much. For instance, the statement "The stability and electro-optical conversion efficiency are better than traditional DWP scheme, whereas the slop efficiency is decreased, and the lasing threshold is increased" is a very weak statement. This is because the statement implies that the slop efficiency has decreased while the lasing threshold has increased. Instead of making a general comment, authors should include the particular data that can be used to compare the performances of the items. If it is at all possible, I would want to encourage the authors to add a table that compares the performance of the currently available core-pumped laser with the proposed clad pump laser.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
A new pumping scheme for 3.5 μm erbium doped fluoride fiber laser is proposed. But the effective implementation of this scheme is under great doubt.
Questions and remarks:
The introduction does not provide requirements for laser pump sources in terms of optical power and divergence parameters, nor does it describe the fiber refractive index profile. The reasons why the conventional pumping scheme is used and the 1976-nm cladding pumping is not used are not specified.
The same rate equations are given as in [16], so please also give the model parameters.
Equation (18) has no indexed variable.
Please, replace “irons” with “ions”.
The reference is wrong: “The mode overlap of a gaussian beam is given by [16]”, the right one, for example, is D. Marcuse, “Gaussian approximation of the fundamental modes of graded-index fibers,” J. Opt. Soc. Am., vol. 68, no. 2, p. 103, Aug. 1978.)
Please, make a figure with illustrative schematic of the proposed fiber pumping and 3.5um fiber laser.
GaSb based diode lasers with 100-um-aperture-width demonstrate about 0.7W of reliable power in CW [12], but the 3.5um fiber laser requires 10-20W. Please, estimate the coupling efficiency.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comment son Numerical Analysis of Dual-Wavelength-Clad-Pumped 3.5 µm Erbium Doped Fluoride Fiber Laser
The paper brings to the readers attention a new method for generating DWP based 3.5 µm lasing using pumping into the fiber cladding only. This investigation is of interest to the field and relevant. However, there are a few important matters that need to be addressed before the manuscript can be published in its current form. The main issues are: 1. The lack of comparison between ordinary (DWP) and clad pumped (DWCP) when both laser operate at the same output power 2. Following from the previous point, the claims for better efficiency of DWCP compared with ordinary DWP that is not substantiated in the current manuscript and is difficult to see whether it is actually true based on the data provided.
Other comments in order of appearance in the manuscript and referred to the line numbering I the provided PDF are provided below:
1. Abstract – Remove comment on higher efficiency unless you can prove it in the manuscript
2. Pg 1, line 25 - 976nm can be core pumped as well, as in Ref [5]
3. Pg. 1, line 26-27 - High beam quality (as in M^2~1) is not mandatory for core pumping with 1976nm since the core is often slightly multimode at this wavelength. It is true that you cannot use ordinary high power multimode diodes.
4. Pg 1. Line 31 - Please provide power levels of the current state-of-the-art 2 µm diodes so the reader can evaluate their commercial maturity compared with thulium fiber lasers.
5. Pg. 3-4 - The rate equations used seems fairly standard for the numerical analysis of 3.5 µm lasers. Unless the authors are using something new in this numerical analysis I suggest referring to the references that already contain similar equations ([15,16] are the most relevant reference) or at least move them to supplementary material
6. Section 3 (results and discussion) – A comparison of DWP and DWCP with the same laser output power (not just the same pump power) is the most important missing data to be able to make a true comparison of the DWCP behaviour and efficiency. I suggest the authors add new versions of the figures as the current figures (which are relevant) but for the same output power.
7. Figures 3,4,5,6,7 – using a 5W P1 (976nm) for simulating is too low for the P2 (1976nm) used because it will result in quenching in almost all cases. It is clear that the authors wanted to demonstrate the quenching effect, however for a true comparison they need to also show the unquenched case of DWP. So a 10 or even 20W of P1 is necessary for this case.
8. In all simulation cases, are we assuming a 16.5/240x260 fiber geometry? When changing core or cladding diameters, is the rest of the fiber cross-section remains the same? Is it getting larger/smaller proportionally?
9. Figure 3 – It would be better to separate these two figures ((a) and (b)) instead of showing then one on top of the other.
10. Figure 3(b) –
a. What is WPE? I assume Wall plug efficiency? Please state and explain clearly how it is calculated and what assumptions are being used (diode efficiency etc.)
b. Is the scale here in actual percentage, i.e., the efficiency is about half of one percent? If so how did you get to these values?
c. I assume we are talking about 1 mol% doping. Does the efficiency improves with higher doping?
11. Pg. 5, line 139 – The authors claim that quenching is “not a thing to be worried currently”. In order to prove this point, again, they need to provide comparative data showing the DWP and DWCP operating at the same output power levels.
12. Figure 4 (and throughout the work) – Some comments on the residual pump levels are necessary. There is clearly more residual pump under DWCP.
Specifically for figure 4. (c), it would be fair to show both sections of the figure with the same Y scale.
13. Figure5, missing units on the right axis
14. Pg 7. What happens if you simulate going to 100, or 200 W of 1973nm? Would quenching show under these conditions?
15. Figure 7 –
a. What core/clad sizes are assumed here
a. At what point along the fibre is the temperature measured?
16. Where in the fiber is the temperature that is shown in Figure 7( c)
17. Pg. 9 , line 232 – “irons” should be “ions”
18. Pg. 9, line 243, in the “conclusions”
a. There is a mention of better stability. I don’t see how this numerical simulation can prove that, even though it might be true in practice, but I don’t think anything in this paper is based on experiments, is that correct?
b. Stating that the electro-optical conversion efficiency. Need to explain what they mean by EO efficiency and how it is calculated (do that not in the conclusion, though), and show conclusive evidence for this claim.
19. Pg. 10 – references – I suggest the authors update reference 1 to the more recent review on the topic by Jackson DOI: 10.1364/OE.400003
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Authors note in the response that "Typical parameters of pumps used in 3.5 μm fiber laser [Zhipeng Qin, Guoqiang Xie, Jingui Ma, Peng Yuan, and Liejia Qian, "Mid-infrared Er:ZBLAN fiber laser reaching 3.68 μm wavelength," Chin. Opt. Lett. 15, 111402- (2017)] are: 976 nm diode with a pigtail fiber of 105 μm core diameter and 0.2 NA; 1976 nm Tm3+:SiO2 fiber with a diameter of 10/130 μm and NA of 0.15. Er3+:ZBLAN fiber used here have a core diameter of 16.5 μm and NA of 0.15, clad diameter of 240x260 μm (double D shape) and NA of 0.5. The refractive index of the ZBLAN fiber core is 1.483.
The reasons why core pumping is used for 1976 nm pump is all about absorption rate. For a 16.5/240x260 fiber, the absorption rate of 1976 nm pump at core pump scheme is over 100 times higher than at clad pump scheme, as shown in figure 5 (c)." But the introduction of the revised manuscript does not include these comments.
Please, see the equation (A7) on line 265. There is nothing under the summation sign. Thus equation (A7) still does not have indexed variables.
Please, check the English in the corrections, for example, "Because most 1976 nm lasers are transporting in fiber clad".
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
I am pleased to see that the authors found my comments useful and followed through them. I think the paper is much improved now and I recommend it is published. One thing that needs to be fixed prior to publication is to re-calculate the heating of the fibre (Fig 7(c)) because I believe the authors used a wrong cladding diameter - see response comment for point 17. Otherwise, my response comments to the various points are below, only minor things.
Point 1: It definitely helps that you add more comparable power levels between DWP and CDWP
Point 2: Noted
Point 3: Noted
Point 4: Noted
Point 5: Noted
Point 6: Thanks
Point 7: Noted
Point 8: Makes sense. Also I suggest adding label “DWP” and “DWCP” on the subfigures themselves, not only in the caption. This will make it easier to see at a glance what do they refer to. Do the same for other figures where one figure is for DWP and the other for DWCP.
Point 9: the point I was trying to convey is that with 5W of P1 there aren't enough ions to be pumped by P2 in most of the figures presented so the behaviour will be non-linear (quenching). In fact, even the example provided here for P1=10W on the right if we look at the laser output power for a FIXED fiber length the power increase is linear.
When changing the fiber length (or other parameter), there will be a complex dependency on other parameters (in fiber length case on output coupling fraction, which needs to be changed as well). It is therefore not possible to convey all of the information with a simple 2D graph. I was therefore suggesting to use a more fair comparison. I cannot argue against your choice of figure aesthetics.
Point 10: Note that the true fibre geometry in Ref [15] is actually 16.5/240x260! The authors used 170 um cladding diameter as a fitting parameter to obtain good fit to the experimental data! Note this will affect the fiber temperature calculation. Regardless of this important point, I think it is necessary to mention the fibre geometry and doping used in the simulations.
Point 11: Noted
Point 12: Noted
Point 13: I still think it would be useful to see a simulation where a DWCP system is pumped to ~600W and see that the quenching is observed as well. Although this is not currently realistic, you can do it in simulation. And the main point would be to see whether the quenching power levels scale in a similar way to the core/clad area ratios.
Point 14: Noted
Point 15: Noted
Point 16: Noted, see point 13. I think it is worth showing Maybe one day 600W diodes will be available and metal coated ZBLAN will be available off the shelf...
Point 17: See my response to point 10. Sorry, I don't see this information provided in the figure. Please write it explicitly.
More important, however, is that the temperature calculation of 7(c) is incorrect because the true cladding diameter is 240x260 um and not 170 um. You also want to mention the diameter of the polymer around the fibre. This is also relevant to the temperature calculation.
Point 18: Noted, see my response to the previous point, 17
Point 19: Noted
Point 20: Noted. I still think it is best to remove the word "stability" here since this is a numerical simulation paper without any experimental results. In a future experimental paper you can properly discuss this point.
Point 21: Noted
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
