Measurement of Ion Mobilities for the Ion-TPC of NvDEx Experiment

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this paper, the authors present a study of the mobility of two different negative ion species in SF6, as a starting point for the development of a 82SeF6 TPC for the NnuDEx experiment. The authors exposed a small TPC to a pulsed laser beam, measured the drift velocities for different electric fields, and tested the possibility of reconstructing the position along the drift direction (z) from the drift time difference of two ion species. Unfortunately, all measurements were taken with the beam at a fixed distance from the readout, making them sensitive to systematic uncertainties and preventing a rigorous check of the z reconstruction. However, this study is a relevant step toward the construction of the NnuDEx TPC, making the paper worth to be published, with an adequate delineation of its real scope.

More detailed comments can be found below.

Title:
- "gas properties" in the title looks too generic, since no other property than the drift velocity is measured. I suggest making the title more precise. An option is to place more emphasis on the observation and exploitation of the two different ion species.

line 16: 
— There is a typo within parentheses, I presume you wanted $0\nu\beta\beta$ here. Same at line 19, 25 and 152.
— idea —> ideal

line 25:
— using high —> using a high

line 33:
— position —> positive

Sec. 2.1:
— Since you don’t have multiplication, it shouldn’t be too difficult to determine the average number of collected ions per beam pulse. Could you provide this information? At least, you should specify if there are just a few or many clusters per pulse. It is relevant to understand if your interpretation of the waveforms is correct (see below).

Sec. 2.2:
— Please quote also the rise time of the CSA response, which is also relevant for the correctness of the waveform analysis (see below). 

Eq.1:
- This equation is correct only if the response of the CSA has a very fast rise time (compared to the I(t) signal development), followed by an exponential decay. Consequently, the rise time should be quoted and a comment on the typical time scale of the signal development is needed to make clear that this equation can be used in your case.

Fig. 5:
— Please add a legend or a note in the caption referring to the colors in the picture

line 104:
— Please provide an estimate of the laser beam width

line 120:
— I’m wondering if mu is indeed the correct estimator for the drift time. The gaussian shape results from at least four contributions: the signal formation in the anode, the electronics response, the laser beam width, and the diffusion. The last two are only relevant if many ions per beam pulse are produced. Using mu is correct only if the laser beam width and/or diffusion are largely dominant over the others. Is it the case? Moreover, if the response of the CSA has some shaping effect on the ms scale, you can cut part of the signal (ballistic loss), resulting in a smaller mu than expected. You could largely remove biases coming from these effects by taking measurements at different drift distances (you would only need to measure how much the shape moves, irrespective of the shape itself), but since you measure only at one position, a correct understanding of the signal shape is needed.

line 132-133:
— The uncertainty associated with pressure and temperature should be better motivated. 2% in the temperature is ~6 degrees. Do you observe or suspect such large fluctuations or inaccuracies in the temperature of the gas? What is the nominal accuracy of your pressure sensors?

line 147-149 and line 159-160:
— I think that these sentences can be easily misunderstood. First, I think that having the distribution centered at 3.7 cm is a direct consequence of assuming 3.7 cm center when you measure the drift velocities. Assuming a 3.8 cm drift distance, you would get this distribution centered at 3.8 cm. So, it tells nothing about the accuracy of the measurement, as the sentence seems to suggest. Again, taking measurements at different drift distances would have provided stronger evidence of the reconstruction capabilities. Second, I guess that the width of the distribution is so small because, for each pulse, you are averaging over many ions. So, it tells nothing about the resolution you will have in your experiment, unless you quote the average number of ions per pulse here and the average number of ions you will have when searching for the neutrinoless double-beta decay. As presented now, the result in Fig. 5 looks no more than a sanity check. I think you should rephrase these sentences to make explicit what can really be extrapolated from your results.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper presents a demonstration of z-reconstuction in SF6 gas at near atmospheric pressure making use of different arrival times for two negative ion species induced by laser pulses, SF6- and SF5-. The mobilities of the two charge carriers were measured. The work is similar to the study in the paper’s Ref [25] done at lower gas pressure, but demonstrates that z-reconstruction in SF6 using this method is possible also at atmospheric pressure. This is an important early step in the realization of the NvDEx experiment, which aims to search for neutrinoless double beta decay in a different but structurally similar gas (SeF6) at even higher pressure.

 

The technical details of the measurement are well described, but I have a small number of comments.

 

In Figure 1 the purpose of the iseg module is not described. I assume this is for high voltage.

 

In Figure 6 there are two orange/red colors that are hard to distinguish.

 

From figure 8 it appears there are approximately 100 laser pulses analyzed for each drift field. Is that correct? Can this be clarified.

Comments on the Quality of English Language

There are a few grammatical errors in the text that should be corrected, and the symbolic representation of neutrinoless double beta decay in my version oddly shows as 0 () fifi presumably due to a typesetting error. Similarly for ohm.

Author Response

Please see the newest attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

I have just comments on two figures, with an addition a list of minor suggested corrections (mostly typos).

Main comments:
1) Figure 4: at line 109, the author write about the long tails, probably due to other heavy ions. But in my opinion some comments on the electronics are due, since they appear as undershoots..... A vision of signals on a higher time range would be useful.
2) Figure 5: The waveforms are fitted with a double gaussian, but the two gaussians are reported separately (and I agree that it is helpful). But it is not appreciable how the double gaussian fits well the waveforms. At least a chi-square information should be given, or put also the sum of the gaussians in the plot.

Minor comments:
1) line 16:is an ideal way
2) line 18: help to study the absolute masses of the neutrinos and their origin.
3) line 33: Both negative and positive ions
4) line 56: turns the current signal into a voltage signal

5) Figure 1: iseg EHS 80-60n is not described in the text (HV system)
6) line 64: they are connected by 10 Mohm resistors.
7) line 109: There is also a long tail
8) line 113: measurement at low pressure
9) line 143: using the same datasets of Section 3
10) line 162: will be improved and extended

Reference 12 not correctly written

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear authors,

          I really appreciates the efforts the authors made to address my concerns. I'm even more convinced of the quality of the results and I support the publication of the paper. I've only a few minor comments:

• Figure 5: I think that using "minority carriers", "majority carriers", "all carriers" instead of "gaussian 1", "gaussian 2", "double gaussian" would make the legend more clear
• Response 10: In this case, I was mostly concerned about the response of your electronics to long signals than the rise time. If your electronics has a shaping effect, it will cut part of your waveform at large times, resulting in a displacement of the peak toward smaller times. If you have indications that it is not the case, it could be worth to mention it in the paper. I don't think it impacts the merit of the paper, but I recommend to take such effects into account in future studies, where the accuracy on drift velocity or signal shape could be more relevant.
• line 168-169: "The closure test...were demonstrated"--> "A closure test ... was successfully performed."

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf