Pith. sign in

REVIEW 4 major objections 5 minor 21 references

A review on development of eco-friendly filters in Nepal for use in cigarettes and masks and Air Pollution Analysis with Machine Learning and SHAP Interpretability

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims CatBoost predicts Nepal's AQI with perfect R-squared scores while a biodegradable HA filter removes over 98% of PM2.5, uniting data-driven pollution analysis with eco-friendly filtration.

desk verdict A broad filter review wrapped around a disconnected ML section whose AQI claims are contradicted by its own dataset description and evaluation metrics. read the letter →

arxiv 2501.13369 v1 pith:ELS3VXPA submitted 2025-01-23 cs.LG cs.AI

Bishwash Paneru , Biplov Paneru , Tanka Mukhiya , Khem Narayan Poudyal This is my paper · ORCID
classification cs.LGcs.AI
keywords AirQualityIndexCatBoostSHAPbiodegradablefilterPM2.5PM10Nepalcigarettefilters
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper combines a review of filter materials with a machine-learning analysis of air quality, aiming to show that a biodegradable Hydrogen-Alpha (HA) filter can protect people in Nepal from the particulate pollutants that most drive the Air Quality Index. The machine-learning section claims CatBoost outperforms XGBoost, Extra Trees, and Random Forest, with the lowest testing RMSE (0.23) and perfect $R^2$ scores (1.00). SHAP analysis is then used to argue that NowCast Concentration and Raw Concentration are the most important features pushing AQI upward. The filter section reports that the HA filter removes more than 98% of PM2.5 and 99.24% of PM10, making it, in the paper's view, a sustainable defense for face masks and cigarette filters. If these claims hold, the same pollutants identified by SHAP are exactly the ones the biodegradable filter removes, connecting forecasting to a concrete public-health intervention.

What carries the argument

The argument runs on three objects. CatBoost is a gradient-boosting algorithm that builds symmetric decision trees and handles categorical features without manual encoding; the paper uses it, alongside XGBoost, Extra Trees, and Random Forest, to map concentration features to AQI. SHAP is a game-theoretic attribution method that assigns each feature a Shapley value, giving the marginal contribution of NowCast Concentration, Raw Concentration, and Parameter to the model's output. The HA filter is the proposed biodegradable filter whose claimed PM removal efficiency is the physical intervention that follows from the SHAP ranking. Nested cross-validation is the evaluation scheme used to support the generalization claim.

What would settle it

Obtain the actual dataset used for the model and print its target column: if it is GHI (global horizontal irradiance) from the EPA repository, then the reported $R^2=1.00$, RMSE 0.23, and SHAP importances are not about the Air Quality Index at all. Separately, challenge the filter claim by testing an HA filter specimen under a standard particle-filtration protocol with NaCl or oil aerosol; recorded PM2.5 removal below 98% or PM10 removal below 99.24% would falsify the central filter result.

Watch

Extended reading notes

Core claim

The paper's central claim is twofold. On the modeling side, it claims that gradient-boosted trees, specifically CatBoost, can predict the Air Quality Index with near-perfect accuracy (training and testing $R^2=1.00$, testing RMSE 0.23), and that SHAP attributions show NowCast Concentration and Raw Concentration are the decisive features, while the Parameter feature contributes almost nothing. On the materials side, it claims that a Hydrogen-Alpha (HA) biodegradable filter, intended for use in face masks and cigarette filters, removes more than 98% of PM2.5 and 99.24% of PM10 with a low pressure drop and full biodegradation within four weeks in composting soil. The paper presents these as two halves of one solution: the model identifies the pollutants that matter, and the filter removes them, offering an eco-friendly response to Nepal's air-quality and waste problems.

Load-bearing premise

The load-bearing premise is that the dataset described in Section 4.1 (an EPA dataset whose columns were cleaned to predict GHI) is the same dataset used for the reported AQI predictions; the paper never identifies an AQI dataset, its location, or its link to Kathmandu, so if the target variable is not AQI the headline results describe a different problem.

Editorial extensions

If this is right

  • If CatBoost's near-perfect scores hold on real AQI data, authorities in Nepal could issue AQI forecasts from NowCast and Raw Concentration readings with little preprocessing beyond cleaning.
  • The SHAP ranking implies that filtration efforts should target PM2.5 and PM10 concentrations rather than secondary variables, because those are the features that move the model's AQI output.
  • If the HA filter sustains more than 98% PM2.5 and 99.24% PM10 removal in field use, switching from cellulose-acetate cigarette filters and polypropylene masks to biodegradable HA filters would cut both inhalation exposure and plastic waste.
  • A low-pressure-drop, moisture-resistant, compostable filter could be reused across masks and cigarette filters, making protection affordable and sustainable in Nepal.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper's two halves are not actually joined by data: Section 4.1 describes an EPA dataset used to predict GHI (global horizontal irradiance), while the results are reported as AQI predictions for Kathmandu. A reader should treat the $R^2=1.00$ and SHAP importance claims as pertaining to an unidentified target until the training data are disclosed.
  • The HA filter's performance numbers appear to be inherited from the cited biodegradable PBS/chitosan filter rather than from new Hydrogen-Alpha material tests; confirming the composition and independently measuring removal efficiency would be needed before deployment.
  • A straightforward testable extension would be to retrain the same four regressors on a properly identified Nepal AQI dataset and compare SHAP rankings; if NowCast and Raw Concentration remain top features, the filter-design recommendation survives, and if not, the link weakens.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript combines a narrative review of biodegradable cigarette and mask filter technologies with a machine-learning section that claims to predict the Air Quality Index (AQI) for Nepal. Four regressors (XGBoost, CatBoost, Extra Trees, Random Forest) are evaluated, with CatBoost reported as best (testing RMSE 0.23, R2 1.00); SHAP analysis is used to claim that NowCast and Raw Concentration dominate AQI predictions. The paper also promotes a biodegradable "HA" filter with more than 98% PM2.5 and 99.24% PM10 removal efficiency and discusses public-health and waste-management implications for Nepal.

Significance. If the ML results were valid, the paper would offer a useful AQI forecasting benchmark for Nepal and an interpretability link to filter design; the review also compiles relevant background literature on filter materials. However, the central ML claims are not supported as reported: the dataset is described as an EPA GHI-prediction dataset, the reported test and nested-CV errors are irreconcilable, and the features used are concentration inputs from which AQI is deterministically derived. The filter section contains no new experimental evidence and appears to restate Choi et al. (2021) under a new name. No code, data, or preprocessing details are provided, so the quantitative results are not reproducible.

major comments (4)
  1. [§4.1 and §5] Section 4.1 states that the dataset was "obtained from an extracted from Environmental Protection Agency (EPA's) repository" and was used "to predict GHI," with 32,151 rows and 4 columns; the surrounding text also mentions DNI and DHI. The abstract and Section 5 report AQI predictions for Nepal from NowCast Concentration and Raw Concentration, but no AQI dataset, monitoring station, date range, or target-variable definition is identified. Because the fitted target may be GHI rather than AQI, the reported R2=1.00, RMSE=0.23, and SHAP importance results do not establish any claim about AQI.
  2. [§5, ML modeling results] The model results table reports CatBoost testing RMSE 0.23 and testing R2 1.00, while the nested cross-validation table immediately below reports mean RMSE 13.85 with standard deviation 2.34. A roughly 60-fold discrepancy cannot describe the same model, target, and evaluation protocol, and the manuscript offers no reconciliation. This directly contradicts the abstract's claim of "greater accuracy and generalization ... cross validated using a nested cross validation approach."
  3. [§5, SHAP results] NowCast Concentration and Raw Concentration are pollutant-concentration inputs from which AQI is computed by a deterministic piecewise-linear mapping. Regressing AQI on these inputs makes R2≈1.00 and their SHAP dominance expected by construction. Section 5's conclusion that high NowCast values "significantly raise" AQI is therefore a restatement of the AQI definition rather than an empirical finding that justifies the HA filter.
  4. [§5 and Table 14] The abbreviation "HA" is never defined, and no experimental procedure, material characterization, or test result for a "Hydrogen-Alpha" filter is presented. Table 14 and Figures 9-10 reuse the removal efficiencies and pressure-drop values (98.3% PM2.5, 99.24% PM10, 59 Pa) from the Choi et al. (2021) PBS/chitosan filter. As written, the manuscript's central filter claim is an attribution to a different named filter, so the conclusion's statement that "the HA filter offers exceptional defense" is unsupported.
minor comments (5)
  1. [§4.1/§5] Section numbering is inconsistent: "4.1 Dataset collection" appears after the "5. Machine Learning approach" heading, and the results section is also numbered 5, which makes it difficult to trace the methods.
  2. [§4.3] Equation numbering is duplicated and references do not match: the RMSE is labeled equation (1) but the text calls it equation (2), the R² formula is labeled equation (3), and the cumulative dose formula is also labeled equation (3). Equations should be renumbered and cited consistently.
  3. [Abstract] The abstract says "AQI is predicted in this work using a Random Forest Regressor," but the results identify CatBoost as the best model; the abstract and conclusion should agree on the primary model.
  4. [Figures] The ML actual-vs-predicted plots and SHAP plots have missing or generic captions such as "Figure : Actual vs predicted plot," and the plots lack axis labels and units.
  5. [References and cited statistics] Some cited statistics appear misreported, for example the Khanal et al. odds-ratio confidence intervals are given with identical ranges across different variables, and the reference list contains incomplete entries (e.g., the Gaussian citation). These should be checked against the original sources.

Circularity Check

1 steps flagged · score 8.0 of 10

The claimed AQI prediction reduces to a deterministic function of its own inputs: AQI is computed from NowCast/Raw Concentration, so R2=1.00 and SHAP dominance are forced by construction; Section 4.1's 'predict GHI' statement further severs the target from the described dataset.

  1. self definitional [Abstract and Section 5 (SHAP results), with dataset description in Section 4.1]
    "NowCast Concentration and Raw Concentration are the most important elements influencing AQI values, according to SHAP research, which shows that the machine learning results are highly accurate."

    The paper predicts AQI using features named NowCast Concentration and Raw Concentration from the EPA repository named in Section 4.1. In EPA methodology, AQI is not an independent outcome: it is computed by a deterministic piecewise function of pollutant concentrations, with NowCast as the PM concentration-averaging input. A regressor fed those concentrations can therefore achieve Testing R2 = 1.00 and RMSE = 0.23 by learning the AQI formula, and SHAP will trivially rank NowCast/Raw Concentration as dominant. The abstract presents this as CatBoost's 'greater accuracy and generalization' and Section 5 uses it to justify the HA filter, but the 'prediction' is equivalent to the target's own definition, not an independent finding.

full rationale

The central ML claim is circular if the target is AQI: the features NowCast Concentration and Raw Concentration are the very quantities from which EPA AQI is defined, so a model trained on them should achieve near-perfect fit and should show those features as overwhelmingly important. The paper exhibits exactly that result and then uses it as evidence of model accuracy and as a motivation for the HA filter. Section 4.1 deepens the problem by stating the EPA dataset was used 'to predict GHI' rather than AQI, meaning the paper never identifies an actual AQI target, station, or date range; the reported testing RMSE of 0.23 also conflicts with the nested cross-validation mean RMSE of 13.85 reported in the same section. The filter sections rely on external measurements from Choi et al. (2021) and are not circular, and the one self-citation (Donato et al., 2023) is not load-bearing for the main claims. However, the flagship ML result is not an independent prediction: it is a tautological fit to a target defined by its own inputs, so the overall circularity score is 8.

Assumptions & free parameters 1 free parameters · 3 assumptions · 1 invented entities

The main load-bearing assumptions are the identity of the dataset (GHI versus AQI), the deterministic relation between AQI and its concentration features, and the unstated identification of the abstract's 'HA' filter with Choi et al.'s filter. No physical constants or model parameters are derived in this paper; the ML hyperparameters are user-chosen.

free parameters (1)
  • ML hyperparameters for XGBoost, CatBoost, and Random Forest = n_estimators=500/200, learning_rate=0.1, max_depth=6/10, etc.
    Chosen by the authors and reported without a tuning search or sensitivity analysis, so their contribution to the claimed R2=1.00 is unquantified.
assumptions (3)
  • domain assumption AQI is a deterministic transform of NowCast and Raw Concentration.
    The SHAP discussion in Section 5 treats NowCast and Raw Concentration as inputs that largely determine AQI, which makes high R2 expected and feature importance partly tautological.
  • ad hoc to paper The EPA dataset described in Section 4.1 as being used for GHI prediction is the same dataset used for the AQI predictions in Section 5.
    Section 4.1 says the dataset was used to predict GHI and that time and sensor cleaning columns were removed, but the abstract and results report AQI predictions without reconciling the mismatch.
  • ad hoc to paper The abstract's 'HA' biodegradable filter is the Choi et al. (2021) PBS/chitosan filter.
    Table 14 and the results section attribute PM2.5 and PM10 removal efficiencies and a 59 Pa pressure drop to Choi et al., but the paper never defines an 'HA' filter, its composition, or its fabrication.
invented entities (1)
  • Hydrogen-Alpha (HA) biodegradable filter
    purpose: Presented as a novel filter for masks and cigarettes that removes over 98% of PM2.5 and 99.24% of PM10.
    The paper provides no synthesis, composition, or test data for an 'HA' filter; the efficiencies and pressure drop are quoted from Choi et al. (2021) without connecting the name, so there is no falsifiable handle specific to this paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A review on development of eco-friendly filters in Nepal for use in cigarettes and masks and Air Pollution Analysis with Machine Learning and SHAP Interpretability." pith.science (2026). https://pith.science/paper/ELS3VXPA

@misc{pith2026250113369,
  author       = {Pith},
  title        = {Pith review of: A review on development of eco-friendly filters in Nepal for use in cigarettes and masks and Air Pollution Analysis with Machine Learning and SHAP Interpretability},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ELS3VXPA}},
  note         = {Machine review of arXiv:2501.13369}
}
read the original abstract

In Nepal, air pollution is a serious public health concern, especially in cities like Kathmandu where particulate matter (PM2.5 and PM10) has a major influence on respiratory health and air quality. The Air Quality Index (AQI) is predicted in this work using a Random Forest Regressor, and the model's predictions are interpreted using SHAP (SHapley Additive exPlanations) analysis. With the lowest Testing RMSE (0.23) and flawless R2 scores (1.00), CatBoost performs better than other models, demonstrating its greater accuracy and generalization which is cross validated using a nested cross validation approach. NowCast Concentration and Raw Concentration are the most important elements influencing AQI values, according to SHAP research, which shows that the machine learning results are highly accurate. Their significance as major contributors to air pollution is highlighted by the fact that high values of these characteristics significantly raise the AQI. This study investigates the Hydrogen-Alpha (HA) biodegradable filter as a novel way to reduce the related health hazards. With removal efficiency of more than 98% for PM2.5 and 99.24% for PM10, the HA filter offers exceptional defense against dangerous airborne particles. These devices, which are biodegradable face masks and cigarette filters, address the environmental issues associated with traditional filters' non-biodegradable trash while also lowering exposure to air contaminants.

Figures

Figures reproduced from arXiv: 2501.13369 by the authors.

Figure 3
Figure 3. Functioning of face masks (Gonzalez et al., 2021). The purpose of this review is to compare the processes and discover the most appropriate process of making safer cigarette and mask filter in developing country like Nepal using modern technologies using past studies and inventions. This review critically analyzes features, functions, and properties of various filters and evaluates potential of such filters for appl… view at source ↗
Figure 7
Figure 7. Curve to represent Cigarettes, cigars, and tobacco consumption in Nepal till 2029. [PITH_FULL_IMAGE:figures/full_fig_p026_7.png] view at source ↗
Figure 8
Figure 8. Bar graph to represent Cigarettes, cigars, and tobacco consumption in Nepal till 2029. [PITH_FULL_IMAGE:figures/full_fig_p027_8.png] view at source ↗
Figures from the paper (2 more)
Figure 9
Figure 9. Figure 9: Scatter plots to show effect of exposure of PM2.5 and PM10 concentration of 98.3% and [PITH_FULL_IMAGE:figures/full_fig_p039_9.png]
Figure 10
Figure 10. Figure 10: Dosage on body as a result of varying parameters. [PITH_FULL_IMAGE:figures/full_fig_p040_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 17 canonical work pages

  1. [1]

    A., Shaheen, B

    Abbas, W. A., Shaheen, B. S., Ghanem, L. G., Badawy, I. M., Abodouh, M. M., Abdou, S. M., Zada, S., & Allam, N. K. (2021). Cost-Effective Face Mask Filter Based on Hybrid Composite Nanofibrous Layers with High Filtration Efficiency. Langmuir, 37(24), 7492 –7502. https://doi.org/10.1021/acs.langmuir.1c00926 Acharya, A. (2024). The Political Ecology of Urba...

    arXiv 2021

  2. [6]

    (2023, March 15)

    Cleveland Clinic. (2023, March 15). Vital Signs . Cleveland Clinic. https://my.clevelandclinic.org/health/articles/10881-vital-signs Cohen, J. E., Krishnan-Sarin, S., Eissenberg, T., Gould, T. J., Berman, M. L., Bhatnagar, A., ... & O’Connor, R. (2022). Balancing Risks and Benefits of E -Cigarettes in the Real World. American journal of public health, 112...

    arXiv 2022

  3. [8]

    Arif, M., & Hassan, E

    Apparatus and method for the production of filters for cigarettes or other smoking articles. Arif, M., & Hassan, E. (2023). Beyond the Haze: A Study of Smog, Its Types, and Mitigation. https://fcci.com.pk/fcci-pdf/E-Book%20Smog.pdf Arredondo, J., Chernyavsky, A. I., & Grando, S. A. (2006). The nicotinic receptor antagonists abolish pathobiologic effects o...

    work page 2023

  4. [9]

    https://doi.org/10.1155/2023/1330946 Kim, Jong, Yeol., Kim, Soo, Ho., Rhee, Moon, Soo., Park, Jin, Won., Kim, Yonug, Hoi. (2009)

    work page doi:10.1155/2023/1330946 2009

  5. [10]

    Model-Based Compensation of Moving Tissue for State Recognition in Robotic-Assisted Pedicle Drilling

    Cigarette filters and cigarettes, comprising paper filters treated with flavoring natural botanical extracts. Konstantinou, E., Fotopoulou, F., Drosos, A., Dimakopoulou, N., Zagoriti, Z., Niarchos, A., ... & Poulas, K. (2018). Tobacco -specific nitrosamines: A literature review. Food and chemical toxicology, 118, 198-203. Konstantinou, E., Fotopoulou, F.,...

    work page Pith review arXiv 2018

  6. [12]

    https://doi.org/10.3390/polym13020207 Martí, M., Tuñón-Molina, A., Finn Lillelund Aachmann, Muramoto, Y., Noda, T., Takayama, K., & Ángel Serrano-Aroca. (2020). Protective face mask filter capable of inactivating SARS- CoV-2, and methicillin -resistant Staphylococcus au reus and Staphylococcus epidermidis. BioRxiv (Cold Spring Harbor Laboratory). https://...

    work page doi:10.3390/polym13020207 2020

  7. [13]

    E., Quintana, P

    Matt, G. E., Quintana, P. J., Destaillats, H., Gundel, L. A., Sleiman, M., Singer, B. C., ... & Hovell, M. F. (2011). Thirdhand tobacco smoke: emerging evidence and arguments for a multidisciplinary research agenda. Environmental health perspectives, 119(9), 1218-1226. Matteucci, M., & Napolitano, R. S. (2015). Apparatus and method for making filters. McG...

    work page 2011

  8. [21]

    Talcott, J

    Filter including electrostatically charged fiber material. Talcott, J. A., Thurber, W. A., Kantor, A. F., Gaensler, E. A., Danahy, J. F., Antman, K. H., & Li, F. P. (1989). Asbestos -associated diseases in a cohort of cigarette -filter workers. New England Journal of Medicine, 321(18), 1220-1223. Tanaka, O., Inui, K., Shibuya, Y., & Kiyotani, H. (2002). M...

    work page 1989

Show all 21 references
  1. [39]

    P., O'Connor, R

    57 Stitt, J. P., O'Connor, R. J., & Kozlowski, L. T. (2003, September). An image processing and analysis systems for automatic classification of cigarette filter blockage. In Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biolog...

    2024

  2. [113]

    Catel-Ferreira, M., Tnani, H., Hellio, C., Cosette, P., & Lebrun, L. (2015). Antiviral effects of polyphenols: development of bio -based cleaning wipes and filters. Journal of virological methods, 212, 1–7. https://doi.org/10.1016/j.jviromet.2014.10.008 Chan, W., Guo, W., & Ji...

    2015 doi

  3. [318]

    & Connolly, G.N

    Hoffmann, D.H., Djordjevic, M.V., Fan, J., Zang, E., Glynn, T. & Connolly, G.N. (1995) Five leading U.S. commercial brands of moist snuff in 1994: Assessment of carci nogenic N - nitrosamines. J. natl Cancer Inst., 87, 1862–1869 Honda, A., Miyazaki, T., Ikegami, T., Iwamoto, J...

    1995

  4. [737]

    A., Kaya, S., & Lin, Y

    https://doi.org/10.3390/su14020737 Singh, A., Dayu, X., Ituen, E., Ansari, K., Quraishi, M. A., Kaya, S., & Lin, Y. (2020). Tobacco extracted from the discarded cigarettes as an inhibitor of copper and zinc corrosion in an ASTM standard D1141 -98(2013) artificial seawater solu...

    2020 doi

  5. [969]

    Martínez, L., Ros, G., & Nieto, G. (2018). Hydroxytyrosol: Health benefits and use as functional ingredient in meat. Medicines, 5(1),

    2018

  6. [1217]

    https://doi.org/10.3390/antiox11071217 Russo, Alessandro. (2021). Filter for cigarettes made with biological material. Sadabad, F. B. (2019). Functional Nanocoatings Inspired by Natural Polyphenols (Doctoral dissertation, Karlsruher Institut für Technologie (KIT)). Santos-Eche...

    2021

  7. [2006]

    Trends Biotechnol

    Laccases, blue enzymes for green chemistry. Trends Biotechnol. 24, 219–226. Roditis, M. L., & Halpern-Felsher, B. (2015). Adolescents' perceptions of risks and benefits of conventional cigarettes, e -cigarettes, and marijuana: a qualitative analysis. Journal of Adolescent Heal...

    2015

  8. [2011]

    New material for airborne viruses filtration. Chem. Eng. J. 173, 341–351. Tobacco Products - Nepal | Statista Market Forecast . (2024). Statista. https://www.statista.com/outlook/cmo/tobacco-products/nepal#revenue Tomkins, A. A., Dulai, G., Dulai, R., Rassenberg, S., Lawless, ...

    2024

  9. [2023]

    Cigarette Filters, Tobacco Tactics, updated 13 December 2021, accessed 03 January

    2021

  10. [2157]

    Cao, X., Zhu, Y., Peng, S., Ouyang, G., Yang, G., Hou, N., Xu, F., & Yan, F. (2023). The influence of forming factors on physical characteristics of hollow filter rod using water vapor forming method. Journal of Materials Research and Technology , 24, 795 –807. https://doi.org...

    2023 doi

  11. [2194]

    Moscatelli, Romano. (2010). A filtering mask. Murphy, J., Gaca, M., Lowe, F., Minet, E., Breheny, D., Prasad, K., ... & Proctor, C. (2017). Assessing modified risk tobacco and nicotine products: description of the scientific framework and assessment of a closed modular electro...

    2010

  12. [3363]

    https://doi.org/10.3390/ma13153363 Pandit, P., Maity, S., Singha, K., Annu, Uzun, M., Shekh, M., & Ahmed, S. (2021). Potential biodegradable face mask to counter environmental impact of Covid -19. Cleaner Engineering and Technology, 4, 100218. https://doi.org/10.1016/j.clet.20...

    2021

  13. [7876]

    Gao, Lin., Wang, Tao. (2017). Manufacturing method of filtering material. Gao, L., Chen, H., & Ren, D. (2019). Preparation method of filter material. Glendening, E.D., Reed, A.E., Carpenter, J.E. and Weinhold, F. (2003) NBO Version 3.1. Gaussian Inc., Pittsburgh. Gonzalez, A.,...

    2017

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.