the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Oct 2024
STRAS: a new high time resolution aerosol sampler for PIXE analysis
Abstract. The joint use of hourly resolution sampling and analyses with accelerated ion beams such as Particle Induced X-ray Emission (PIXE) technique has allowed the measurement of hourly temporal patterns of particulate matter (PM) composition at many sites in different parts of the world. The demand within the scientific community for this type of analysis has been continuously increasing in recent years, but hourly resolution samplers suitable for PIXE analysis are now discontinued and/or suffer from some technical limitations. In this framework, a new hourly sampler, STRAS (Size and Time Resolved Aerosol Sampler), was developed for the collection of PM10, PM2.5 or PM1. It allows automatic sequential sampling of up to 168 hourly samples (1 week), it is mechanically robust, compact, and easily transportable. To increase PIXE sensitivity, each sample is concentrated on a small surface area on a polycarbonate membrane. The comparison between the elemental concentrations retrieved by STRAS samples and samples collected using a standard sequential sampler operated in parallel shows a very good agreement; indeed, if both the samplers use the same kind of membrane, the concentrations of all detected elements are in agreement within 10 %.
- Preprint
(2604 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on amt-2024-137', Anonymous Referee #1, 11 Nov 2024
The manuscript titled “STRAS: a new high time resolution aerosol sampler for PIXE analysis” has developed a new hourly sampler, called STRAS, which can be used to automatically collect sequential samples of up to 168. Specially, a small surface area for depositing particles is designed for detecting elements using the Particle Induced X-ray Emission (PIXE) technique. Field measurements were conducted to perform comparison between STRAS and other sampling instruments. Overall, it is a good advancement of the sampling technique. There are several comments that need to be addressed as below.
The main issue is how to verify the accuracy and reproductivity of the newly invented sampler? The manuscript has compared the results between STRAS and Gemini, which both shared the same inlet, cabinet, pump, air flow control system as stated by the authors. As samplers are subject to various uncertainties and Gemini is not a reference sampler, the good consistency between STRAS and Gemini doesn’t guarantee the reliability of STRAS. It is essential to use standard particles with known concentrations and compositions for verifying a new instrument.
Section 4.1, Since the authors have revealed the filter collection efficiency, is this collection efficiency robust or random? how to account for the loss of particles in the application?
Line 95 - 105: The description is not written in a scientific manner, it is more like a manual.
Line 288 – 289: I cannot foresee the application of STRAS in measurement of black carbon or brown carbon as aethalometer has advantages especially on its time resolution.
Citation: https://doi.org/10.5194/amt-2024-137-RC1 -
RC2: 'Comment on amt-2024-137', Anonymous Referee #3, 24 Nov 2024
General comments:
This paper introduces a new aerosol sampler for observing particulate matter composition capable of taking many samples with high temporal resolution, offering the potential for more efficient atmospheric monitoring—an increasingly critical need. However, the manuscript lacks several key elements, including a quantitative evaluation and thorough discussion of the sampling method, as well as a comparison of its advantages and disadvantages relative to conventional methods. These points require further elaboration.
A major concern is the observed loss of fine particles for certain elements compared to Teflon filters. This raises questions about the suitability of the method for fine particle sampling (PM2.5, PM1). If the sampler is intended for fine particle applications, the authors should clearly discuss its validity and propose solutions to address these issues.
Additionally, it was difficult to understand the superiority of STRAS over the conventional streaker sampler and the Gemini sampler used for this comparison. I expect that STRAS allows the collection of a larger number of samples and a larger amounts of particles in one spot, but there is insufficient data regarding the Germini sampler’s capacity, making it difficult for readers to draw meaningful comparisons. I would recommend to make a table to explain merits and limitations of each sampling method together with other methods mentioned in the introduction.
Specific comments
- Introduction:
・As noted in the general comments, please consider adding a table to summarize the merits and limitations of different sampling and analytical methods. In addition, include additional details, such as the differences in detection limits, applicable elements, costs, etc., between EDXRF and PIXE, and the advantages of STRAS sampling system over traditional streaker samplers. This would enhance the clarity of the introduction.
- Sampler design:
・Please include information about the size and weight of STRAS sampler, since you mentioned that the sampler was designed to be “compact and easily transportable.”
・Please provide details on how particle size resolution is achieved. There was no information on size separation. In addition, it was not clear for me whether the sampler can collect different size fractions at a same time or not.
- PIXE analysis of STRAS samples
・P6. L174: How did you determine the errors (especially for deposit area)? If they were based on analytical data or references, include this information.
・P6. L177: If filters were washed with acid, would this reduce the blank values? Trace metal concentrations can be as low as several ng/m3 in remote areas like marine atmosphere.
- STRAS validation
・P7. L184: Please specify the sampling position (longitude, latitude) and height.
・4.1: You mentioned potential underestimation of S and K due to their enrichment in fine particles. However, you don’t address any solutions or recommendations for it. In addition, I don’t understand why other elements have good agreement even in PM1 (Fig. 7) despite that they are all fine particles.
・4.2: Please explain why Germini data has larger errors (for example, Figure 5, 6). Is this due to the smaller sample quantities on the filter spots?
・Figure 7: Why Ca concentrations at low levels are higher in STRAS sampler compared to the Germini sampler?
- Conclusion
・P12 L284: This should be mentioned in section 4. Also please refer some papers here.
Technical comments
・P2. L46: optimal solution ⇒ optimal analytical solution
・P3. L93-94: I don’t think the sentence “Furthermore…samplers.” is necessary in the introduction, this should be moved to conclusion or discussion.
・P8. L224: How about adding “(e.g. S and K)” at the end of the sentence?
Citation: https://doi.org/10.5194/amt-2024-137-RC2 -
RC3: 'Comment on amt-2024-137', Anonymous Referee #4, 02 Dec 2024
This manuscript describes the STRASS equipment, an aerosol sampler for the collection of atmospheric particles. Sampling is performed in short periods of one hour, over a polycarbonate filter surface; that permits the collection of 168 samples, sequentially and automatically, during one week. The exposed filter is sequentially transported to the laboratory, where the direct measurement of elementary aerosol composition by PIXE is performed, without further treatment. This sampling/PIXE analysis methodology permits the evaluation of trace aerosol composition with hourly discrimination at acceptable costs and is therefore potentially interesting for source apportionment purposes.
The STRASS sampler was built to replace and improve the previous STREAKER sampler used for the same objectives, which is not commercially available, anymore.
In the paper the characteristics and performance of the STRASS sampler are evaluated with various experiments to validate the sampler and polycarbonate 0.8 µm pore membrane capability to collect correctly PM1, PM2.5 and PM10 aerosols.
The evaluation of 0.8 µm pore polycarbonate filter was done by parallel sampling with a Teflon membrane filter recognized as having high filter efficiency for submicrometric particles. The results, exposed in Figure 3, show very similar concentration values. However, the presented results are for PM10 and elements usually associated with coarse particles (with possible exception of S). Therefore, from the experiment, it is not clearly demonstrated the capability of the 0.8 µm pore membrane to collect efficiently fine particles. Results for elements principally associated with fine particles and for PM1 or PM2.5 fractions would be more relevant to the evaluation of filter collection efficiency.
The reliability and accuracy of the STRASS sampler was done in a field validation, by parallel sampling in the external environment with the STRASS and a GEMINI sampler, using both the same polycarbonate 0.8 µm pore membranes. The results show remarkable similarities for both samplers, principally for PM10, a clear indication of the reliability of the STRASS sampler which seems not to suffer from flow control or leak problems.
The collection capability of 0.8 µm pore membrane filters is further evaluated in field conditions by using parallel sampling with the GEMINI sampler provided with Teflon membrane filters. The results presented in Figure7 for PM1 particles show evidences of inefficient polycarbonate membrane filtration for elements concentrated in submicrometric size ranges, such as S and K. Contrary to previous figures, figure 7 shows scatterplots with X and Y in logarithmic scales. From the figure it seems that presented linear regressions and correlation coefficients are also for the logarithm of concentrations. This presentation permits the evidencing of lower concentration values, that, anyway, are already strongly influenced by PIXE detection limits and filter blanks variability. However, visualization of differences at higher concentration ranges are reduced. A parallel figure using linear scales would be useful for a more clear evaluation of polycarbonate filters performance.
Citation: https://doi.org/10.5194/amt-2024-137-RC3
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 156 | 32 | 27 | 215 | 5 | 2 |
- HTML: 156
- PDF: 32
- XML: 27
- Total: 215
- BibTeX: 5
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1