This manuscript describes the comparison of the performance between the “High-Performance” HP-SPAMS and the prior version (SPAMS). While limited design details are provided, data is included for the scattering efficiency and hit fraction vs size for both instruments, as well as mass spectral resolution. Yet, there are many general statements in the manuscript for which details/data are not provided, or for which the authors cite a manuscript for a different single-particle mass spectrometer. Line numbers referred to in this review correspond to the track changes version of the manuscript.
As noted in the previous reviews, details are missing from the Section 2 description of the HP-SPAMS. Detailed descriptions of the specific parts that we changed (including both the previous and current parts) are needed. Some details were added in the revision, and this is helpful. Yet, it still is often not clear what exactly changed between the two versions of the instrument, and most of the statements are vague/general. The improvements made to the instrument appear to include adding an aerodynamic particle concentrator system, applying delayed ion extraction, using a high frequency YAG laser, and adding four-channel data acquisition.
The authors have previously published focused manuscripts about the design and characterization of the two main improvements to instrument:
1) Delayed ion extraction:
- Li et al. 2018, J. Am. Soc. Mass Spec., “Improvement in the Mass Resolution of Single Particle Mass Spectrometry using Delayed Ion Extraction”
- Chen et al. 2020, Atmos. Meas. Tech., “Increase of the particle hit rate in a laser single-particle mass spectrometer by pulse delayed extraction technology”
2) Improved aerosol inlet:
- Du et al. 2023, Atmosphere, “Design and simulation of aerosol inlet system for particulate matter with a wide size range”
While Li et al. (2018) and Chen et al. (2020) are cited, it is not clear that that details of the design and previous characterization can be found in those manuscripts, and it would be very helpful for the current manuscript for this to be made clear in Section 2. Further, these previous manuscripts provided data on hit %s and mass spectral resolution that could be used in the current manuscript to comment on how much each improvement made a difference to the overall performance of the HP-SPAMS.
Based on these previous manuscripts, the authors’ main goal in the current manuscript appears to be the characterization of the full integrated system (HP-SPAMS) and comparison to the original SPAMS, rather than a detailed description of the “design”, which is lacking, as noted above. The manuscript title needs to be revised to indicate this.
Several of the improvements made to this instrument have been previously described for other single-particle mass spectrometers, but this is not clear in the manuscript. This comment relates to the inadequate response of the authors to the previous requests for additional comparisons to non-SPAMS instruments. I provide examples (not meant to be comprehensive – the authors need to do a literature search) here. Delayed pulse extraction is used by several single-particle mass specs: miniSPLAT (Zelenyuk et al., 2015, J. Am. Soc. Mass Spec.), BAMS (Czerwieniec et al. 2005, J. Am. Soc. Mass Spec.), SPASS (Erdmann et al. 2005, Aerosol Sci. Technol.), LAMPAS-2 (Costa Vera et al. 2005, Rapid. Commun. Mass Spectrom.), ALABAMA (Clemen et al. 2020, AMT). Four-channel data acquisition is used on other single-particle mass spectrometers, including the ALABAMA (Brands et al. 2011, Aerosol Sci. Technol.) and ATOFMS (Pratt et al. 2009, Analytical Chem.). The 100 Hz Nd:YAG laser (Centurion, Quantel) is used on the Univ. of Michigan ATOFMS (Gunsch et al. 2017, Atmos. Chem. Phys.).
The authors should cite Su et al. (2023, Atmos. Environ., “Analysis performance of single particle aerosol mass spectrometer for accurate sizing and isotopic analysis of individual particles”) as an additional study that characterized the SPAMS. In particular, that study also reported the accuracy of the 208Pb/206Pb ratio to be closer than observed for the SPAMS in the current work (Figure 10 in the current manuscript). This comparison should be addressed. In addition, it is not only the correlation that matters (discussed on Line 291 and shown in Figure 10), but also the comparison to the theoretical isotope ratio, which is not provided in the current manuscript.
The authors state on Lines 38-39: “However, the limitations of current mass spectrometry (MS) detection capabilities render it not well suited for analyzing complex aerosol components in low concentration environments.” This is not accurate, as there have been single-particle MS measurements in the Arctic, free troposphere, and stratosphere – all low concentration environments. Only two references are cited in the next sentence, but there are many papers – especially using PALMS, SPLAT, and ATOFMS – that have shown highly successful data obtained in these environments.
On Lines 42-43 the authors refer to needing long-time collection and changing mixing state, but this is confusing because SPMS is, by definition, a real-time analysis method, such that long-time collection is not possible. Further, Pratt et al. (2009, Analytical. Chem.) shows aerosol mixing state with 2 and 4-min resolution. Therefore, it is clear that the authors’ statement needs revision for multiple reasons.
The authors state on Lines 44-45: “…the poor mass resolution and detection sensitivity of SPMS make it difficult to obtain an accurate analysis of aerosols. (Pratt et al. 2009; Zelenyuk et al. 2009)” Yet, neither of these papers shows this, and it is not an accurate statement, as there are many publications, especially using ATOFMS, SPLAT, and PALMS showing accurate, quantitative analysis by SPMS.
On Lines 56-57, the authors state “The particles that could be transmitted to the instrument are not entirely measurable, which could be related to the particle sizing efficiency and ionization probability”, with no reference. This statement is confusing, as many published single-particle mass spec characterization studies have quantified transmission through multiple methods.
On Lines 80-81, the authors state “Although there are many studies to improve one aspect of the performance of SPMS, the instrument’s overall performance is insufficient.” Yet, there have been many improvements to single-particle mass spec instruments over the past three decades, with great success, and many high-performing instruments exist. Please revise this statement to refer to the need to improve the SPAMS itself. Further, this statement makes it clear that a paragraph about the SPAMS itself is needed in Section 1, as the main goal of the current work is to compare the SPAMS and HP-SPAMS.
Line 94: Provide the height with the inlet included as well.
Line 95: Provide the actual, rather than approximate weight.
On Lines 101-102, it is stated “…most particles have entered the separation cone.” Please provide data, and quantify “most”.
Lines 106-107 state “…could theoretically transport particles in the range of 100-5000 nm.” Where is the theoretical transmission curve data? Further, scattering efficiency data appears to only go out to ~2-3 um in Figure 4.
Lines 112-113 state “Compared to the SPAMS, increasing the laser power enhanced the intensity of the scattered light from the particles.” Please state how much the laser power was increased by, as well as provide data comparing the scattered light.
Lines 114-115 state “the background noise level was effectively reduced, and the detection capability for small-size particles was improved.” This is referred to again on Lines 190-192 with the same phrasing. How much was the noise level reduced? What size is “small”? How much was the detection of these “small-size” particles improved? Provide quantitative data.
Line 125 states that “laser beam homogenization” is employed, but I could not find a description of how this was done. The reference (Steele et al. 2005) is to a different single-particle mass spectrometer. Please clarify how this was set-up for the HP-SPAMS and that is an improvement over the SPAMS. Also, it is confusing when, on Lines 212-213, the authors refer to a “non-uniform Gaussian beam”, suggesting that laser beam homogenization may not have been done? But, then Lines 255-256 mention a “uniform laser beam”. So, the laser set-up is not clear.
Lines 161-162: It is stated “To compare the detection capability of the instrument at low concentrations, the fraction of the samples fed by the SPAMS and HP-SPAMS was diluted by 40 times…”. Please provide the initial and diluted number concentrations, to provide important context for what is meant by “low”.
Lines 189-190: Please provide the current and previous critical orifice sizes for context. The reference here to Cahill et al. (2014) is for a different single-particle mass spectrometer.
Lines 210-212: Here it is stated that particles of different composition require different laser energy thresholds, but hit % data is only provided for the ambient data (Figure 5) and is not provided for the four different particle types shown in Figure 4, for which chemically-dependent scattering efficiency data are shown. Add hit % data that corresponds to the lab scattering data in Figure 4, as presumably the entire instrument was running during the experiment, such that hit data should be available. Increase the discussion of chemical biases here, including the role of the low chosen laser power (0.5 mJ).
Report the HP-SPAMS data in terms of number concentration, rather than #/min, in Figure 6. Then in the text, compare to the SMPS number concentration in the same size range, to the HP-SPAMS number concentration, to enable a quantitative comparison. This is important because the authors are claiming very high scattering and detection efficiencies in Figures 4-5, and so, a quantitative comparison is needed here. Currently only correlation analysis was completed, at the bottom of page 7. It would also be more meaningful to examine the correlation with the SMPS concentration in the same size range where the HP-SPAMS has high scattering and detection efficiency. The scatter and hit rates of the instrument are also important to discuss to provide context for the instrument performance.
Line 235-236: I could not find where these data are provided. Please clarify what is meant by “detect” (hit?), and provide data. Is this an average value or a maximum?
Lines 251-256: This discussion suggests that the ion transmission above m/z 100 improved, or the detection of these ions improved. Page 9 points to the data acquisition board change (signal amplification), which makes sense, but it isn’t mentioned here as a likely explanation.
Please clarify the sentence on Lines 278-279. Are the authors referring in the second part to the increased detection of ions > m/z 100? Also, how does this factor of 47.8 compare to the expected based on the data in Figures 4 and 5?
Additional comments:
- Line 16: By “multiple homogeneous masses”, do the authors mean “isobaric ions”?
- Line 28: Incomplete sentence
- Line 30: Change “outstanding” to “improved”
- Lines 33-34: Add reference to a review.
- Line 108: Is one laser used with its beam split, or are two lasers used? Please clarify.
- Lines 116-117: I assume that authors are referring to the calibration of velocity to dva. Please clarify the description here.
- Line 170: There appears to be a typo or English phrasing problem here.
- Line 205: I assume the authors mean “delayed ion extraction” rather than direct current extraction here?
- Line 238: Do the authors mean “nuclei” instead of “nuclear”?
- Lines 247-249: The authors should cite their previous paper here, as it showed a detailed analysis of this improved resolution and would support the statement made.
- Lines 260-261: This statement is confusing. Please clarify.
- Line 273: Please add “compared to the SPAMS” to provide context for the statement “more sensitive and accurate”.
- Line 290: Include figure # here.
- Lines 296-278: Clarify in this sentence that these improvements are compared to the SPAMS.
- Line 307: This states 0.97 corresponds to R-square, but Line 221 states it is a Pearson correlation (i.e. r, not r2). Which is correct?
- Lines 311-312: It is stated “although the sensitivity of HP-SPAMS to analyze individual particles cannot be quantified”. I don’t understand this statement, as this is what the authors have characterized in this paper.
- Line 314: Why is this not attributed to the ion detection, rather than the ion generation?
- Figure 4: Please change “Detection Efficiency” to “Scattering Efficiency” on the plot y axes to match the caption revision.
- Figure 6: Change “variation curve” to “temporal plot”. Is the DUST METER the PM2.5 concentration? Provide the size range of the SMPS for context. Explain the HP-SPAMS plot legend in the figure caption.
- Figure 9: The c & d x axis labels show “m/z (Da)”, but the figure caption suggests that the plot is showing m/z 206 relative peak area. Please fix.
- Line 496: Fix to 208Pb+ vs 206Pb+ to agree with figure. |
