Fiber Optic All-Polarization Weak Magnetic Field Sensor Based on Sagnac Interferometer

Round 1

Reviewer 1 Report

The article reported by Liang et al. entitled ”Fiber optic all-polarization weak magnetic field sensor based on Signac interferometer" demonstrates the experimental realization of magnetic field sensor based on SI. The proposed method and approach are correct to the best of my knowledge. However, some open points need to be addressed before any consideration. Comments are given below: 

1. State-of-the-art must be improved. The introduction is not enough and brief to show the need for the device and how it overcomes the challenges and limitations of previous devices. 

2. More detail must be included on SI and sensing principle

3. What is the rule of optimization?

4. How many times did the experiment was repeated? What was the maximum deviation observed? Authors are advised to show an error bar in Fig . 4.

5. What other parameters were considered during the examination? How to avoid cross sensitivity?

6. Is the response reversible or not? The author should vary the MF from increasing to decreasing and from decreasing to increasing and show the hysteresis plot.

7. dispersed typo and grammatical errors.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Fiber optic magnetic field sensor is important and interesting field currently. The authors proposed the fiber optic sensor for magnetic field based on the Sagnac interferometer scheme. The paper must be improved.

1. The Sagnac interferometer is well-known scheme and there are a lot of works where it was used for electrical current (i.e. magnetic field) detection. So the authors should mention about it in introduction and give the brief explanation what is new in their variant.

2. The eq.1 and 4 should be discussed and explained. The authors should explain why  they use such combination of the Jones matrices for circuit description.

3. All matrices (C, Y, T) should be written with explanation.

4. The authors write that epsilon is extinction ratio of the polarizer. There is no polariser on the figure 1.

5. At the entrance to the interferometer there is IOC, but there are no description of it in the text

6. Due to magnetic rotation at the sensing element there will be cross-talk between modes of the fiber. The eq.1 does not  take this into account.

7. The authors uses ASE as a source. They should write the spectrum width and the power of the source. It is also should be explained why they use broadband source instead of the monochromatic one.

8. At the figure 4 it is not understandable what is the unit of the Output data. If it is an arbitrary unit then they should be normalised to 1 and not to be in the range 8e9 to 2e10.

9. At the figure 4.b there are two lines "before fitting" and "after fitting". It should be explained in the text what is the "fitting" and why the curve became more linear after "fitting"

10.The authors make the brief discussion of the noise sources in the proposed scheme. They should mention what kind of noise was main in their experiments.

11. There are two main parameters of the interference schemes: the sensitivity of the optical scheme to the phase difference of the waves and the transformation coefficient from the external influence (the magnetic field in this study) to the phase difference. The authors should mention this two values.

Author Response

The general comments: Fiber optic magnetic field sensor is important and interesting field currently. The authors proposed the fiber optic sensor for magnetic field based on the Sagnac interferometer scheme. The paper must be improved.

 

Our response: We thank the reviewer for the questions and comments for helping us significantly improve our manuscript. We have addressed all the concerns provided by the reviewer.

 

Question 1: The Sagnac interferometer is well-known scheme and there are a lot of works where it was used for electrical current (i.e. magnetic field) detection. So the authors should mention about it in introduction and give the brief explanation what is new in their variant.

 

Our response: We appreciate the reviewer for this comment. In our work, the Sagnac interferometric structure is used for polarization detection, which is quite different from the phase detection in those schemes for electrical current detection. As we all know, the magnetic field directly changes the polarization of the propagation light instead of phase. In those schemes, quarter wave plates are needed to complete the polarization-phase conversion twice, which introduces more errors. We have added the explanation in the introduction part.

 

The new description:

“Aiming at the problems mentioned above, a fiber-optic all-polarization Sagnac magnetic field sensor is proposed. The Sagnac structure is used for polarization detection, making the system more concise than the phase detection schemes which need quarter wave plates. Besides, the common path Sagnac interference makes it possible for high stability measurement.”

 

Question 2: The eq.1 and 4 should be discussed and explained. The authors should explain why they use such combination of the Jones matrices for circuit description.

 

Our response: We appreciate the reviewer for this suggestion. In the polarization optical system, the characteristics of the optical device can be described by the Jones matrix, or the Mueller matrix. Usually, when it comes to solving the problem of partially polarized light, the Mueller matrix can be used, and the Mueller matrix is related to the intensity of the optical signal; when it comes to the problem of polarized light interference, the Jones matrix can be used, and the Jones matrix is related to the polarization amplitude and phase of light. According to the characteristics we concerned, the Jones matrix is chosen to establish the theoretical model. The order of the matrix is the reverse order of light propagating through the devices.

 

Question 3: All matrices (C, Y, T) should be written with explanation.

 

Our response: We appreciate the reviewer for this suggestion. We have added the explanation of these matrices in this revision.

 

The new description:

“where C represents the coupler, Y represents the IOC, T represents the PM fiber, represents the Jones matrix of the sensing unit for CW light”

 

Question 4: The authors write that epsilon is extinction ratio of the polarizer. There is no polariser on the figure 1.

 

Our response: We appreciate the reviewer for this comment. The integrated optical chip (IOC) is composed of a polarizer and a Y-branch. The extinction ratio here refers to the extinction ratio of IOC.

 

Question 5: At the entrance to the interferometer there is IOC, but there are no description of it in the text.

 

Our response: We appreciate the reviewer for this comment. We have added the description of it in this revision.

 

The new description:

“The optical signal enters the IOC which is composed of a polarizer and a Y-branch after passing through the coupler and is divided into two signals, which propagate in the clockwise (CW) and counter-clockwise (CCW) directions respectively.”

 

 

Question 6: Due to magnetic rotation at the sensing element there will be cross-talk between modes of the fiber. The eq.1 does not take this into account. 

 

Our response: We appreciate the reviewer for this suggestion. Eq.1 is a simplified theoretical model indeed, and we don’t take some factors into account to make it clearer to understand the basic sensing principle. As for the cross talk between modes, we set up an all-polarization system and decrease the length of fiber to minimize the influence of it.

 

Question 7: The authors uses ASE as a source. They should write the spectrum width and the power of the source. It is also should be explained why they use broadband source instead of the monochromatic one.

 

Our response: We appreciate the reviewer for this suggestion. The reason of using broadband source is to reduce the adverse effects of back reflections and backscattering in Sagnac interferometric system. We have added the parameters of the source in this revision.

 

The new description:

“The light source is an amplified spontaneous emission (ASE) broad-band source with a working wavelength of 1550 nm, a spectrum width of 40 nm, and a power of 80 mW.”

 

 

Question 8: At the figure 4 it is not understandable what is the unit of the Output data. If it is an arbitrary unit then they should be normalised to 1 and not to be in the range 8e9 to 2e10.

 

Our response: We appreciate the reviewer for this comment. The date of Y axis is the digital quantity of the system output, which can be converted to magnetic field by a scaling factor.

 

 

Question 9: At the figure 4.b there are two lines "before fitting" and "after fitting". It should be explained in the text what is the "fitting" and why the curve became more linear after "fitting"

 

Our response: We appreciate the reviewer for this comment. “fitting” here means data fitting. The “before fitting” curve is the actual data we acquired from the system and the “after fitting” curve is the result of first-order fitting that we performed on the “before fitting” data. By comparing these two curves, the linearity of the system can be calculated.

Question 10: The authors make the brief discussion of the noise sources in the proposed scheme. They should mention what kind of noise was main in their experiments.

 

Our response: We appreciate the reviewer for this suggestion. The weights of different noises are kind of hard to specify, as we have trouble in decomposing the final error into individual noises. To the best of my knowledge, the optical noise should play the major role.

 

Question 11: There are two main parameters of the interference schemes: the sensitivity of the optical scheme to the phase difference of the waves and the transformation coefficient from the external influence (the magnetic field in this study) to the phase difference. The authors should mention this two values.

 

Our response: We appreciate the reviewer for this suggestion. The sensitivity of the optical scheme to the phase difference is decided on the Sagnac system itself, which is about 1e-7 rad. The transformation coefficient from the external influence here essentially depends on the magneto-optic crystal, which is about 8400 rad/m/T. We have added the descriptions in this revision.

 

The new description:

“The magneto-optical crystal is a cylindrical Ga:YIG crystal with a coefficient of 8400 rad/m/T, a length of 2.1 mm, and a diameter of 1.8 mm.”

 

Round 2

Reviewer 1 Report

 

Authors provided satisfactory response of almost every concern and now the manuscript is in acceptable format. However I still feel that the author do not understand my question number 4 about error bar.

It would be nice if author can show the error bar in Fig. 4 (b). It is highly necessary to see how stable their sensor and sensing response is. Author may find following article useful in order to plot the error bar. 10.3390/jsan11030046 (Fig. 9),10.3390/polym14102110 (Fig. 9 d,e,f),10.3390/s21072403 (Fig. 8), etc.

 

Author Response

We appreciate the reviewer for the further suggestion and we have added the error bar in Fig.4(b).

Reviewer 2 Report

The authors have made good revision of the text. It can be published now.

Author Response

We appreciate the reviewer so much for the comments and helping us significantly improve the manuscript.