Retrieval of the Depolarization Ratio of Graphite Particles Using an Aerosol Chamber

Park, Gahyeon; Kim, Dukhyeon; Noh, Youngmin; Cho, Gangnam

doi:10.5194/egusphere-2026-2900

Preprints

https://doi.org/10.5194/egusphere-2026-2900

Preprints

10 Aug 2026

| 10 Aug 2026

Retrieval of the Depolarization Ratio of Graphite Particles Using an Aerosol Chamber

Gahyeon Park, Dukhyeon Kim, Youngmin Noh, and Gangnam Cho

Abstract. This study derives the intrinsic depolarization ratio of graphite particles using a custom-built near-field polarization light detection and ranging (LiDAR) system combined with an aerosol chamber. To minimize the inherent geometric overlap problem in typical LiDAR systems, the receiver and chamber were separated by 30 m and signals were acquired at ultrahigh spatial resolution (0.03 m). Pure graphite particles and fugitive graphite particles collected from a steel plant were individually injected into a standardized chamber using a dust feeder. Parallel and perpendicular polarized backscatter signals were then measured at a wavelength of 532 nm. Background aerosol signals in the acquired raw data were eliminated through signal preprocessing and a correction algorithm. Ultimately, the intrinsic depolarization ratio was calculated as 0.15 ± 0.01 for pure graphite and 0.16 ± 0.03 for mixed particles collected from the steel plant. The results are statistically consistent (within the margin of error). The derived optical indicators provide a scientific foundation for the future development of remote monitoring technologies that can measure fugitive dust originating from steel plants in real time.

Received: 21 May 2026 – Discussion started: 10 Aug 2026

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Gahyeon Park, Dukhyeon Kim, Youngmin Noh, and Gangnam Cho

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-2900', Anonymous Referee #1, 26 Aug 2026
The authors present measurements of the particle linear depolarisation ratio of graphite particles performed with a novel experimental setup that combines a lidar system with an aerosol chamber. In principle, the work is of interest to the readers of AMT. However, there are a number of issues that have to be addressed in a major revision before this work could be considered for publication.
Major issues:
I strongly recommend that the authors consider using community-approved names and signs for the considered parameters as it is not always clear what they are referring to. This matters as the depolarisation ratio is a parameter that demands very specific information. Please consult Freudenthaler et al. (2009, https://doi.org/10.1111/j.1600-0889.2008.00396.x) as a standard reference. In my understanding, the total mixed depolarisation ratio is what's known as volume linear depolarisation ratio and the intrinsic depolarisation ratio is the particle linear depolarisation ratio (PLDR). In any case I didn't double-check the equations in Section 2.3 because of the irritating nomenclature.

I am missing a discussion of relevance. Specifically, I would be interested about the size range and concentration of fugitive dust from steel plants. Don't they have measures for depositing most of this material? Is whatever is actually emitted comparable to the size distribution and concentration generated in the aerosol chamber? This matters for assessing the use of the chamber measurements for atmospheric application.

I generally like the setup that the authors have developed for this study. However, I would actually expect that a novel setup is tested in a series of measurements to assess performance: Does an empty chamber reproduce the PLDR of the outside air? Does a measurement with spherical particles give a PLDR of zero? Can you reproduce literature values when feeding mineral dust into the chamber? It's somewhat audacious to present new findings with an unverified setup.

Minor issues:
line 40: a PLDR of unity means that the reflected light is completely unpolarised and this doesn't really happen. Note that PLDR gives an indication of sphericity or nonsphericity and that "completely nonspherical particles" is a meaningless designation.

Section 2: I suggest to remove subsubsections and restructure as 2.1 Experimental setup, 2.2 Lidar instrument, 2.3 PLDR calculation

Figure 2 could be omitted. The only relevant information is the distance between lidar and chamber (which could be added to Figure 1) and that measurements were performed outside.

line 108: Is this an open chamber or are there windows on the sides that could affect the depolarisation measurement? Do you have any measurements of particle size distributions or concentration? Have you checked that the particles are properly mobilised and homogeneously distributed in the chamber?

Line130 and subsequent equations: I am missing the effect of the molecules in the air. The mixed depolarisation ratio (if it's the same as VLDR) should include molecules, graphite, and all the other ambient aerosols. The latter two contribute to PLDR which then needs to be disentangles for graphite and the rest.

Line 138 and line 150: Applying a background value and a correction factor seems a lot like you are doing a relative calibration of the PLDR. If that's the case, please state it so that readers understand what you are doing. Also, couldn't you perform an absolute calibration of the PLDR?

Figures 4 and 5: Not sure if both are needed. There is certainly too little text to warrant both.

Table 2: explain the meaning of the parameters in the table header.

Figure 7: This is nice to see but the information is the same as in Figure 8. So maybe omit or move to appendix?

Figure 8: What do the error bars represent? Are they the standard deviation of the mean of the values in Figure 7? If that's the case, do you also have an estimate of the absolute error? An absolute calibration of the PLDR or measurements along the lines of major comment 3 could be of help here.
Citation: https://doi.org/10.5194/egusphere-2026-2900-RC1
RC2: 'Comment on egusphere-2026-2900', Anonymous Referee #2, 03 Oct 2026

Review of

Retrieval of the Depolarization Ratio of Graphite Particles Using an Aerosol Chamber

by Gahyeon Park et al.

egusphere-2026-2900

The scientific quality of the manuscript is insufficient and more than major corrections should to be done.

The major outcome of the described experiment is promised to be the "precise measurement" of "the unique optical properties of graphite fugitive dust emitted from steelmaking processes", but only the linear depolarization ratio is addressed and this in a scientifically insufficient manner.

The measurement setup is described insufficiently.

Main possible error sources are not considered.

The analysis method is not appropriate.
I propose to reject the manuscript.

Details:
Several possible sources of depolarisation in the received lidar signal other than the backscattering from the graphite aerosol are not considered.

Although the 532 nm output of a NdYAG laser after the second harmonic generator should in principle be very well linearly polarized, an output window of the laser can be birefringent and cause depolarisation. In this respect the laser is insufficiently described.

The 90° prism in the emitter path can add laser beam depolarisation, for example due to birefringence, and the coating of the steering mirror can exhibit diattenuation and retardation, leading possibly to elliptically polarised light, which shows up as depolarization in a linear polarization analyser.

Furthermore, it is not described how accurate the orientation of the laser beams plane of linear polarisation is rotated with respect to the incidence plane of the analysing polarizing beam splitter and how much the rotation uncertainty could contribute to the measured depolarization.
Another possible source of depolarisation is the window of the telescope, if it has one. But the telescope is not described sufficiently.
The divergence of the laser beam is not given. It can be up to 6 mrad full width containing 86% of the energy of an assumed Gaussian beam for the mentioned type of laser. The corresponding diameter of the laser beam at the location of the chamber in 30 m distance can be up to 180 mm and backscattering from the frame of chamber frame can not be excluded, adding to the measured depolarisation. Therefore the description of the experimental setup must contain how accurately the laser beam was aligned in the chamber and how this alignment has been controlled.
The laser beam divergence determines the diameter of the scattering volume in the chamber and together with the diameter of the telescope aperture they determine the maximum incidence angle in the telescope. Considering the angular magnification of the combination of the telescope and the collimation lens, the maximum incidence angle in the interference filter with 1 nm bandwidth and in the polarizing beam splitter can be calculated. The polarization transmission and diattenuation of both are angle dependent. It must be discussed and estimated how much this effects the measured depolarisation ratio.
The used polarizing beam splitter cube (Thorlabs, CCM1-PBS25-532/M) has Tp = >95% and Rs >99.5%, which means that there is considerable, but not well specified cross-talk from the parallel to the cross polarized signal and a bit the other way. This is not considered here.

After accurately measuring the cross talk, this effect could be corrected, but the used theoretical concept does not allow for that.
Moreover, the used theoretical concept of parallel and cross polarized backscatter coefficients is shown to be a wrong concept already by Gary Gimmestadt in 2008 (Gimmestad, G. G.: Reexamination of depolarization in lidar measurements, Appl. Opt., 47, 3795-3802, 2008.)

V. Freudenthaler showed in 2016 how to deal physically and mathematically correct with aerosol backscattering and polarization error sources in lidar. (Freudenthaler, V.: About the effects of polarising optics on lidar signals and the Δ90 calibration, Atmos. Meas. Tech., 9, 4181–4255, https://doi.org/10.5194/amt-9-4181-2016, 2016. )
Neither the detectors nor the pre-amplifiers are specified and there is no information about the linearity of their analogue amplification.

Amplifiers have analogue bandwidths that limit the temporal resolution of a lidar system, which is not specified. Here only the nominal spatial resolution of the ADC, that is the lecroy waverunner oscilloscope, is given. The analogue bandwidth and the length of the laser pulse, which is up to roughly 10 ns or about 1.5 m signal length, are possible reasons for the larger than three meters width of the depolarization features in Fig. 7. But also the specification of the spatial distribution of the graphite aerosols in the chamber is too vague and rather obfuscating with the following sentences:

"...flow rate of the graphite powder was 2 m/s (approximately 2.5 g/h), creating a constant particle concentration in the environment. To prevent the injected particles from settling under gravity, thus maintaining a uniform suspended state over an extended period, the internal air was gently circulated using a fan installed at the top of the chamber. "

The curvature of the top of the pure graphite depolarization ratio profiles (red) in Fig. 7 indicates a change of the graphite particle concentration over the length of the chamber or a spatial smoothing of the peak signal due to analogue bandwidth limitation. Both effects have an impact on the retrieval of the aerosol depolarization ratio from the volume depolarisation ratio using a constant "background aerosol".
The most critical mistake of the correction and retrieval process is probably the calibration of the amplification ratio of the two backscatter signals with an assumed linear depolarisation ratio of 0.03 of the background aerosol, using a constant value retrieved in Finland by Bohlmann et al. , 2021, under probably very different environmental conditions and aerosol sources, for the location in Korea for two different dates, one in June and one in November, with again different environmental conditions. Freudenthaler (2016, see above) shows, how the polarization calibration can be done independently of environmental assumptions.

The other data processing steps described in Fig. 4 and in the text below after line 150 are not clear at all, and the description of the background correction is completely missing.

Furthermore:
The term "near field" is not appropriate for the near range detection with a lidar. The general use of "near field" in contrast to "Fresnel region" and "far field" can be found in https://en.wikipedia.org/wiki/Near_and_far_field or in any appropriate textbook of optics. For the lidar we must use "near range" and "far range" instead.
Line 15:

What does " The results are statistically consistent. " mean?
Line 16 ff:

It is not clear what the meaning of this sentence could be:

"The derived optical indicators provide a scientific foundation for the future development of remote monitoring

technologies that can measure fugitive dust originating from steel plants in real time. "
Line 40 "depolarization ratio":

As there are lidars using circular polarized and and others using linear polarized light emission - and even those that use both in one system, we must discern between circular depolarization ratio and linear depolarization ratio. It should be made clear at least once at the begin of the manuscript, that the "depolarization ratio" used in the manuscript is a shortcut for "linear depolarization ratio" in the rest of the text.
Line 40 ff:

"... the depolarization ratio (D p ), an important indicator of particle nonsphericity. Ranging from 0 (entirely spherical) to 1 (completely nonspherical), ..."

"completely nonspherical" is not a scientific term and not helpful at all for describing the properties of aerosol particles that cause depolarized backscatter. Furthermore, a linear depolarization ratio close to 1 has never been observed with lidar.
Line 46 ff:

"If the D_p of a specific particle is known, the presence and contribution of that particle within mixed aerosols can be estimated

from LiDAR observations."

The method is not described and no reference given.
Line 58:

" Operating organically..."

What does this mean?
Line 66:

"...beam quality and stability. "

What does this mean?
Line 71:

" The collected light was converted to parallel light ..."

This is not possible, because there is always a certain divergence from an extended source.

"Collimation" is the appropriate expression.
Line 73:

" Each separated signal passed through an interference filter..."

This is in contrast to Fig. 1, where only one interference filter is shown before the polarizing beam splitter cube.
Line 102ff:

"Repeated preliminary experiments confirmed that 30 m minimizes the geometric overlap effect of the LiDAR system while securing stable backscattered signals. "

This description lacks any useful information for identifying and possibly criticizing the procedure.

Furthermore, what does "securing stable backscatter signals" mean?
Line 109ff:

" ... ensuring that the laser pulses could generate spatially significant signals within the chamber."

What are "spatially significant signals"?
Line 112:

"...flow rate of the graphite powder was 2 m/s (approximately 2.5 g/h), creating a constant particle concentration in the environment. "

How were a flow rate of 2 m/s, a particle injection of 2.5 g/h and a constant particle concentration determined and controlled together with a "uniform suspended state"?
Line 120ff:

"...degree of particle sphericity D_p, ..."

This does not make sense.

Same sentence:

"...also defined as the ratio of the depolarized backscattered signal to the nondepolarized signal."

This is wrong.

Line 129:

Reference to Klett, 1981 is wrong. This paper does not deal with polarization in lidar signals.
Line 133:

"total mixed depolarization ratio" => this is the usually in the lidar literature used "volume depolarization ratio".

Line 168ff:

" Each step is designed to maximize the data precision and reliability while minimizing the effects of environmental variables through rigorous calibration."

What are the uncertainties resulting from the different processing steps which are "maximizing the precision"?

Which "rigorous" calibration is meant.

Due to lack of knowledge, I cannot comment on the part of the manuscript that deals with the composition and structure of the measured aerosol particles. However, some sentences are not understandable, independent of this knowledge. These are:

Line 199ff:

"Here, the optical properties of the graphite particles were quantified through repeated experiments within a controlled chamber environment, ultimately compiling a robust dataset of 24 experimental sessions to verify statistical reliability"

- Which "optical properties"?

- What are the "repeated experiments"?

- What is the "controlled chamber environment"?

- What does "robust dataset" mean?

- What does "statisical reliability" mean?
Line 218:

"... additional scattering variability caused by surface attachments."

What does this mean?
Line 219:

And what does "statistically consistent" mean?

Citation: https://doi.org/10.5194/egusphere-2026-2900-RC2

Gahyeon Park, Dukhyeon Kim, Youngmin Noh, and Gangnam Cho

Viewed

Total article views: 219 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
116	72	31	219	29	21

HTML: 116
PDF: 72
XML: 31
Total: 219
BibTeX: 29
EndNote: 21

Views and downloads (calculated since 10 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	39	18	6	63
Sep 2026	45	39	18	102
Oct 2026	32	15	7	54

Cumulative views and downloads (calculated since 10 Aug 2026)

Month	HTML	PDF	XML	Total
Aug 2026	39	18	6	63
Sep 2026	45	39	18	102
Oct 2026	32	15	7	54

Viewed (geographical distribution)

Total article views: 181 (including HTML, PDF, and XML) Thereof 181 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 24 Oct 2026

Short summary

Steel plants release harmful graphite dust that is hard to track. To solve this, we used a custom light-based remote sensing system and a controlled chamber to measure how graphite reflects light. We discovered that graphite has a unique optical fingerprint, reflecting light differently than natural dust or other pollution. This distinct signature allows us to remotely identify and track graphite pollution in real time, helping to improve industrial air quality monitoring and public health.


Total:	0
HTML:	0
PDF:	0
XML:	0