Dispersive Fourier Transform Spectrometer Based on Mode-Locked Er-Doped Fiber Laser for Ammonia Sensing

Round 1

Reviewer 1 Report

The submitted manuscript describes a spectrometer based on Dispersive Fourier Transform for ammonia sensing. The proposed method/setup is of significant importance since dispersive Fourier Transform inherits a high potential for fast and broadband spectroscopy. The manuscript has high scientific level and provide a good discussion on the results and their limitations. I recommend the manuscript for publishing.

Here are my comments:

·         In line 155 Authors said: ‘Subsequently, the gas cell was purged with air, reducing the NH3 concentration, and a spectral measurement for the new concentration was conducted.’ My first thought was that the Authors just purged the cell with extra air, increasing the pressure. I suggest to rephrase this sentence so the procedure for lowering the concentration will be more clear. I also suggest including the volume of the cell and the pressure. 

·         It would be useful to include spectra with different concentrations in Fig 4.

·         The variables in equation on line 147 should be described below the equation.

The manuscript should include more description of the importance of the work done and compare it to other systems/works indicating advantages. As it is written now, it is not clear why Authors chose to use DFT for the measurements, since they do not used its advantages such as high speed and bandwidth, while the results do not surpass performance of traditional spectrum analyzer.

The quality of English could be improved in the abstract and Introduction.

Author Response

General comment

The submitted manuscript describes a spectrometer based on Dispersive Fourier Transform for ammonia sensing. The proposed method/setup is of significant importance since dispersive Fourier Transform inherits a high potential for fast and broadband spectroscopy. The manuscript has high scientific level and provide a good discussion on the results and their limitations. I recommend the manuscript for publishing. 

Our response

We thank the Reviewer for the evaluation of our work. 

Comment 1

In line 155 Authors said: ‘Subsequently, the gas cell was purged with air, reducing the NH3 concentration, and a spectral measurement for the new concentration was conducted.’ My first thought was that the Authors just purged the cell with extra air, increasing the pressure. I suggest to rephrase this sentence so the procedure for lowering the concentration will be more clear. I also suggest including the volume of the cell and the pressure.

Our response

We agree with this comment. Indeed, it doesn't sound clear. We have revised the sentence as following:

To assess the DFT spectrometer’s capability for quantitative NH3 detection, we conducted the following experiment: initially, the gas cell was filled with a mixture of NH3 and air, achieving a concentration in the tens of percent range at the atmospheric pressure, and the spectral measurement was performed.

Subsequently, the cell was purged with air to reduce the NH3 concentration while maintaining the atmospheric pressure within. Then, the spectral measurement for the new concentration was conducted. The volume of the cell is 500 cm3.

Comment 2

It would be useful to include spectra with different concentrations in Fig 4.

Our response

We have revised the Fig.4, and added the derived absorption spectra with different concentrations. 

Comment 3

The variables in equation on line 147 should be described below the equation.

Our response

We have added the description of the variables that were not discussed earlier in the manuscript. 

Comment 4

The manuscript should include more description of the importance of the work done and compare it to other systems/works indicating advantages. As it is written now, it is not clear why Authors chose to use DFT for the measurements, since they do not used its advantages such as high speed and bandwidth, while the results do not surpass performance of traditional spectrum analyzer.

Our response

Thanks for bringing this to our attention. The purpose of this work was to demonstrate for the first time the possibility of using DFT measurements to determine the concentration of ammonia, as well as to study the limitations that reduce the maximum sensitivity of the method. In particular, we have shown that even the weak absorption of 10^-20 cm2/molecule exhibited by ammonia near wavelengths of 1530 nm can be used for its detection using conventional erbium-doped mode-locked lasers. The approach is distinguished by its ability to make measurements down to microsecond time-scale. Although the speed of spectroscopic measurement was not an important parameter in the current work, this advantage of the proposed DFT system can be used, for example, in applications related to monitoring reaction rates in chemical plants, where fast acquisition rate is desirable. Another possible application of the approach is the rapid analysis and monitoring of the concentration of typical gaseous biomarkers, such as nitric oxide (NO), ammonia (NH3), and methane (CH4). These biomarkers are actively and relatively abundantly produced in the body, play a significant role in physiological processes, and are released through breath.

We have added this information to the text:

We have shown that even the weak absorption of $10^-20$ cm$^2$/molecule exhibited by ammonia near wavelengths of 1530 nm can be used for its detection using such the conventional erbium-doped mode-locked laser. 

Our estimates indicate that the created DFT scheme is capable of detecting NH$_3$ concentrations as low as 0.1\%. 

Furthermore, we explore the potential for enhancing sensitivity to probe concentrations at the ppm-level. The implemented approach is distinguished by its ability to measure down to microsecond time-scale. Although the speed of spectroscopic measurement was not an important parameter in the current work, this advantage of the proposed DFT system can be used, for example, in applications related to monitoring reaction rates in chemical plants, where fast acquisition rate is desirable. 

Reviewer 2 Report

In this work, the authors report a DFT-spectrometer based on a mode-locked tunable fiber laser. NH3 absorption spectroscopy with a 0.2 nm spectral resolution is demonstrated. Potential avenues for enhancing sensitivity and spectral resolution have also been discussed. The results sound credible and interesting. In my opinion, the manuscript deserves publication with major revision. Some suggestions are given as follows.

1. In Results section, the author said “we averaged spectra acquired from time traces of 6400 consecutive pulses, resulting in an effective time resolution of 0.128 milliseconds”. The question is what is the spectrum like if the average value is not taken.

2. The spectrum shown in Fig. 2 is smooth. However, there are many dips in the spectrum given in Fig. 3. What is the reason for these dips? How to distinguish the absorption dip with other dips?

Author Response

General comment

In this work, the authors report a DFT-spectrometer based on a mode-locked tunable fiber laser. NH3 absorption spectroscopy with a 0.2 nm spectral resolution is demonstrated. Potential avenues for enhancing sensitivity and spectral resolution have also been discussed. The results sound credible and interesting. In my opinion, the manuscript deserves publication with major revision. Some suggestions are given as follows. 

Our response

We thank the Reviewer for the evaluation of our work. 

Comment 1

In Results section, the author said “we averaged spectra acquired from time traces of 6400 consecutive pulses, resulting in an effective time resolution of 0.128 milliseconds”. The question is what is the spectrum like if the average value is not taken.

Our response

The unaveraged spectrum has the worse signal-to-noise ratio, that can be estimated as 15 dB, though it still can be utilized for the concentration estimations. We have added an example of the unaveraged spectrum to Fig.3. We have also mentioned the signal-to-noise ratio in the manuscript as follows:

With the given electrical amplification set in our experiments, the noise level amounts to δU = 2.5 mV RMS, that with U=200 mV gives the signal-to-noise ratio of 19 dB.

Comment 2

The spectrum shown in Fig. 2 is smooth. However, there are many dips in the spectrum given in Fig. 3. What is the reason for these dips? How to distinguish the absorption dip with other dips?

Our response 

Please note that while there is an intense absorption line of ammonia at 1531.6 nm, there are many other lines that fall into the generation spectrum as well (see Fig. 2, green line). Thus, the spectrum, transmitted through the gas cell, acquires many low-prominent dips, each of those potentially possesses information about the ammonia concentration. 

We have added the following sentence into the manuscript:

Note that there are many absorption lines falling into the generation spectrum (as per Fig.2), spectra of the radiation transmitted through the gas appeared to have many low-prominent dips with the most intense appearing at 1531.6 nm. 

Reviewer 3 Report

In this work, the Authors demonstrated a 1.5 μm Er-doped fiber laser system featuring an Dispersive Fourier Transform Spectrometer for ammonia sensing. The authors succeeded in presenting a picosecond pulse generation at a repetition rate of 25 MHz. They provide a sufficiently large amount of experimental data to prove the declared 0.2-nm spectral resolution of the DFT spectrometer.

However, the main drawback of the manuscript is the absence of significant novelty in the presented results. Authors haven’t shown advantages of their method compared to well-known and widely used direct comb-spectroscopy or dual-comb spectroscopy for such tasks. Moreover, authors declared that main noise factor is amplification noise and quantization distortion of oscilloscope limiting minimal concentration measurement by only 0.1% which is crucially low performance compared to typical comb-spectroscopy methods (see e.g. https://doi.org/10.1364/OE.16.002387). In addition, the statement about negligible influence on concentration measurement of jitter and amplitude noise of master oscillator (h.e. RIN performance and linewidth of single comb line) is confusing and is still under the question. Authors should describe these issues more carefully.

Author Response

General comment 

In this work, the Authors demonstrated a 1.5 μm Er-doped fiber laser system featuring an Dispersive Fourier Transform Spectrometer for ammonia sensing. The authors succeeded in presenting a picosecond pulse generation at a repetition rate of 25 MHz. They provide a sufficiently large amount of experimental data to prove the declared 0.2-nm spectral resolution of the DFT spectrometer. 

However, the main drawback of the manuscript is the absence of significant novelty in the presented results. Authors haven’t shown advantages of their method compared to well-known and widely used direct comb-spectroscopy or dual-comb spectroscopy for such tasks. Moreover, authors declared that main noise factor is amplification noise and quantization distortion of oscilloscope limiting minimal concentration measurement by only 0.1% which is crucially low performance compared to typical comb-spectroscopy methods (see e.g. https://doi.org/10.1364/OE.16.002387). In addition, the statement about negligible influence on concentration measurement of jitter and amplitude noise of master oscillator (h.e. RIN performance and linewidth of single comb line) is confusing and is still under the question. Authors should describe these issues more carefully.

Our response

We thank the Reviewer for the evaluation of our work.

Indeed, spectroscopy of optical combos, double combos, as well as other spectroscopy methods make it possible to measure concentrations down to ppb. Thus, in the article https://doi.org/10.1364/OE.16.002387, a sensitivity of up to 18 ppb was obtained in the circuit. It should be noted that such sensitivity is achieved using non-trivial technical approaches, such as adjusting the modes of the resonator-based gas cell and the modes of the optical comb, and in the case of two-comb spectroscopy, the need to use two combs simultaneously. In addition, in the case of cavity-enhanced spectroscopy, the use of a spectral analyzer is necessary. All this complicates the implementation of such approaches.

The DFT method has such advantages as the relative simplicity of its implementation: no adjustment of the optical comb is required, and, that might be even more important, no spectrometer is required to measure the absorption spectrum. Finally, for some applications, such as monitoring intermediate products of chemical reactions in chemical production, measurement speed will play a significant role, and our demonstrated minimum detectable absorption of 10-3 cm-1 with a cell of the length of 20 cm and sampling speed of tens of milliseconds can compete with other proposed methods. For example, in 10.1364/OL.37.003285 it’s proposed to measure methane presence with a time resolution of the order of 100 ms, and a minimum detectable absorption of 10-5 cm-1 (at an equivalent cell length). Potentially, in our work, an increase in sensitivity can be achieved by using a mode-locked laser with a higher peak power.

We have added this consideration to the text as follows:

The achieved resolution and sensitivity open avenues for practical applications, demanding high acquisition rates, such as monitoring intermediate products of chemical reactions in chemical production, where fast and easily to implement proposed realization might be especially desirable.  Potentially, an increase in sensitivity can be achieved by using a mode-locked laser with a higher peak power, as well as by utilizing a multipath gas cell and stronger dispersion line. 

We also thank the Reviewer for mentioning the importance of the laser amplitude noise and jitter. Their impact within the DFT method can be estimated by studying the variations of time-traces of different pulses within the pulse train. We have revised the variations of shape of each pulse and found them to be roughly two times larger than the scope’s noise level. 

We have revised the corresponding part of the discussion as follows:

Firstly, there are noises from the pulse laser, including pulse jitter and amplitude jitter. 

Secondly, the SOA utilized to amplify the signal possesses a gain that also experiences fluctuations.  

These two types of noises are attributed as optical noises. 

Thirdly, there are detector noises, mainly shot noise and thermal noise, contributing at high frequencies. 

Lastly, there are noises from the oscilloscope, including amplification noise and quantization distortion. These two types of noises, attributed to as electrical noises  $\delta U_electrical$, may be estimated with measurements of background noise at the absence of the signal at the scope’s input.  

The joint impact of the optical and electronic noises can be estimated from the statistics acquired for the large pulse train, as it appears at the Fig.~\ref{fig:spectra_comparison}. 

With the given electrical amplification set in our experiments, the electrical noise level amounts to $\delta U_electrical = 1.6$~mV RMS, while the joint impact of both optical and electrical noises yields $\delta U_electrical = 2.5$~mV RMS, that with $U=200 $~mV gives the signal-to-noise ratio of 19 dB.

In this case, the minimum detectable concentration is of the order of 0.1\%, which aligns well with the minimum concentration observed in the experiments (see Figure~\ref{fig:concentration_variation}). Note, that optical noises such as pulse jitter do not have profound effect on the minimal detectable concentration, being only two times larger than the electrical noises.

Reviewer 4 Report

The paper introduces a dispersive Fourier transformation (DFT) spectrometer, which converts spectral information into temporal waveforms. Using a single-pixel photodetector and a pulsed laser enables efficient absorption spectroscopy of ammonia gas with varied concentrations. The designed spectrometer achieves high-resolution detection of NH3 with a spectral resolution of 0.2 nm, employing a mode-locked tunable fiber laser at 1531.6 nm. The study further investigates the system limitations and proposes strategies for heightened sensitivity and spectral precision. This exploration marks a step towards robust and precise gas sensing technologies. A few questions need to be addressed to improve the manuscript.

1.       The authors mentioned that they utilized the HITRAN database for ammonia concentration estimation in the abstract, but they did not discuss the details of the implementation. This could be useful information if added.

2.       The authors discussed the noise level of DFT which limits the spectral resolution. Can the authors quantify the signal-to-noise ratio of their DFT?

Author Response

General comment 

The paper introduces a dispersive Fourier transformation (DFT) spectrometer, which converts spectral information into temporal waveforms. Using a single-pixel photodetector and a pulsed laser enables efficient absorption spectroscopy of ammonia gas with varied concentrations. The designed spectrometer achieves high-resolution detection of NH3 with a spectral resolution of 0.2 nm, employing a mode-locked tunable fiber laser at 1531.6 nm. The study further investigates the system limitations and proposes strategies for heightened sensitivity and spectral precision. This exploration marks a step towards robust and precise gas sensing technologies. A few questions need to be addressed to improve the manuscript.

Our response

We thank the Reviewer for the evaluation of our work. 

Comment 1

The authors mentioned that they utilized the HITRAN database for ammonia concentration estimation in the abstract, but they did not discuss the details of the implementation. This could be useful information if added.

Our response

We have added the calculation details for the absorption spectra used to concentration estimation:

We calculated the ammonia absorbance using the HITRAN database at 0 Celsius degrees, a pressure of 1 atmosphere, without apodization, and using the Hartmann-Tran profile at the negligible concentrations of ammonia. Note that this calculation does not imply resonance collision broadening that may occur at higher NH$_3$ concentration.  That may bring the relative error at the high concentrations, that we estimate as \textbf{1%}.

Comment 2

The authors discussed the noise level of DFT which limits the spectral resolution. Can the authors quantify the signal-to-noise ratio of their DFT?

Our response

Signal-to-noise ratio is defined as δU/ U_0. We have revised the value of  δU, and mentioned the signal-to-noise ratio as follows:

With the given electrical amplification set in our experiments, the noise level amounts to δU = 2.5 mV RMS, that with U=200 mV gives the signal—noise ratio of 19 dB. 

Round 2

Reviewer 2 Report

The author answered the question raised, but it was not detailed. I think the manuscript could be accepted in current version. 

Reviewer 3 Report

I have no additional comments.