the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 Nov 2025
Real‑Time Water‑Soluble Iron Speciation in Ambient Aerosols at Neutral and Slightly Acidic pH
Abstract. We present the first online instrument for the speciation of water‑soluble iron in ambient aerosols, enabling simultaneous quantification of Fe(II) (ws‑Fe(II)) and total water‑soluble Fe (total ws‑Fe). The system combines flow injection analysis with spectrophotometric detection of the Fe(II)–ferrozine complex using a liquid waveguide capillary cell (LWCC) for sensitive detection. The setup was tested with two different aerosol sampling units during field campaigns in Berlin. In summer 2024, the Metrohm AeRosol Sampler (MARS) operated at pH 6.5, while in winter 2025 a particle‑into‑liquid sampler (PILS) was applied at pH 4.5 to mimic acidic cloud water conditions. Limits of quantification (LOQ) for Fe(II) determination were 1.6 ng m⁻³ and 1.0 ng m⁻³ for the MARS‑FIA and PILS‑FIA setups, respectively, with ambient ws‑Fe concentrations ranging from below LOQ to 47 ng m⁻³. Both setups yielded robust online measurements; however, the PILS‑FIA working at pH 4.5 underestimated ws‑Fe compared to filter sampling and extraction. This discrepancy can be attributed to the shorter extraction time in the PILS system, highlighting the influence of extraction duration on the measured iron concentration. Since soluble iron drives important tropospheric aqueous-phase reactions like hydroxyl radical formation through Fenton chemistry, the speciation data provided by the presented setup could improve model representations of atmospheric iron processes.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Aerosol Research.
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.- Preprint
(2559 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 24 Dec 2025)
- RC1: 'Comment on ar-2025-37', Anonymous Referee #1, 28 Nov 2025 reply
-
RC2: 'Comment on ar-2025-37', Anonymous Referee #2, 01 Dec 2025
reply
This manuscript presents the first online instrument capable of simultaneously measuring ws-Fe(II) and total ws-Fe in ambient aerosols using a FIA–LWCC system coupled to two aerosol sampling units (MARS and PILS). The development is technically meaningful, and the instrument shows promising potential for improving the time-resolved quantification of dissolved iron and soluble iron species—key parameters for understanding atmospheric aqueous-phase chemistry and Fe redox cycling.
Before acceptance, however, several scientific and methodological issues require careful clarification and revision. I recommend revision. Detailed comments are provided below.
- Clarification of lines 45–47 — Misinterpretation of literature
Original statement:
“particularly when exposed to solar radiation and oxalate with nanometer sized, amorphous iron, especially from combustion sources (Baldo et al., 2022; Ito et al., 2019).”It is unclear what the authors intend to convey. Do you mean that pyrogenic Fe exhibits higher dissolution rates under acidic conditions? If so, this needs to be stated explicitly.
However, neither Baldo et al. (2022) nor Ito et al. (2019) make the statement currently implied in your sentence.
Baldo et al. found that iron from coal combustion dissolves much faster than mineral dust under simulated atmospheric acidic conditions. Ito et al. emphasized that pyrogenic Fe shows higher solubility relative to crustal Fe.
Please rewrite this sentence to clearly reflect the actual conclusions of the cited studies and your intended message.
- Definition of “water-soluble Fe (ws-Fe)” and discussion of extraction chemistry (Lines 64–66)
You define ws-Fe as Fe extracted either at pH 6.5 (MARS) or at pH 4.5 (PILS), simulating cloud water. Later in the manuscript, ws-Fe appears to be described as the sum of Fe soluble in Milli-Q water plus Fe soluble in slightly acidic solution, which introduces conceptual inconsistency.
Currently, the description contradicts standard definitions. Many previous studies quantify soluble Fe using ammonium acetate (pH ~4.3) to approximate “readily soluble” Fe (e.g., Perron et al., 2020, Talanta, https://doi.org/10.1016/j.talanta.2019.120377).
I strongly recommend adding a subsection in Methods explicitly addressing:
- How this definition compares with commonly used leaching solutions (e.g., acetate buffer).
- How the choice of pH affects the interpretation of “soluble Fe” relative to atmospheric processes.
Clear definition and justification are crucial for interpreting the presented ws-Fe and ws-Fe(II) data.
- Some questions of the online analysis workflow
Time resolution of PILS sampling
“The reduced sample flow also leads to a longer sampling time of 12 min for the PILS system.”
Does this mean the temporal resolution of PILS-FIA measurement is 12 min?
Valve switching during Fe(II) and total Fe measurement
During the sequence in which HA or water is introduced for Fe(III) reduction:
Is the six-port valve switched to waste, meaning no aerosol sample solution is injected during Fe(III) reduction and Fe(II)–FZ complex formation?
If so, this should be clearly illustrated in Figure 2 and explained in the text. This is essential for understanding how continuous the measurements truly are.
- Reduction of Fe(III) by hydroxylamine – residence time and reaction completeness
In the reduction coil, Fe(III) is reduced to Fe(II) by hydroxylamine hydrochloride (HA). Unlike some batch methods using ascorbic acid that require ~30 min residence time to ensure complete Fe(III) reduction (Perron et al., 2020, Talanta; Zhi et al., 2025, ES&T, https://doi.org/10.1021/acs.est.4c12370), the manuscript does not discuss:
- the residence time in the reduction coil,
- whether the mixing with HA and ferrozine in the tubing is sufficient,
- or any calibration tests to confirm complete Fe(III) reduction within the online system.
Please provide:
- the internal volume and corresponding residence time in the reduction coil and reaction coil,
- experimental evidence (or literature support) that HA achieves complete Fe(III) → Fe(II) reduction within this timeframe,
- discussion of potential underestimation of total Fe if reduction is incomplete.
This is a critical issue because incomplete Fe(III) reduction would directly bias the reported total ws-Fe.
- Organic-ligand–driven Fe dissolution
line 445 “They demonstrated that ws-Fe is formed through the dissolution of iron in water mediated by specific organic compounds present in the exhaust.”
The current statement focuses on the large driver in water-soluble iron from primary VOCs in exhaust. Please expand it briefly to reflect that, in addition to primary VOC-derived organics, common atmospheric ligands such as formate, acetate, and oxalate can complex Fe³⁺ and enhance both dissolution and photoreduction to Fe²⁺ (Shi et al., 2012, https://doi.org/10.1016/j.aeolia.2012.03.001; Chen and Grassian, 2013, https://doi.org/10.1021/es401285s). Recent observations also show oxalate enrichment in aged supermicron particles (Li et al. 2025, https://doi.org/10.1093/nsr/nwaf221), implying aging-dependent effects. Please revise this section to incorporate these studies and clarify the organic-mediated pathways promoting Fe solubility.
Citation: https://doi.org/10.5194/ar-2025-37-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 172 | 27 | 19 | 218 | 18 | 22 |
- HTML: 172
- PDF: 27
- XML: 19
- Total: 218
- BibTeX: 18
- EndNote: 22
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This paper presents what I would call preliminary results from the development and brief deployment of two instruments for measuring water soluble Fe (ws Fe) in ambient PM2.5 particles and speciation into Fe(II) and Fe(III). The ambient measurements are over two separate periods with a different particle collection system for each and cover a total of 3 weeks of sampling. Measuring ws Fe is difficult due to low concentrations and the measurement depends on the method operational parameters. Possibly one of the biggest difficulties is measuring what is in the ambient particle since it is difficult to sample the particles unaltered using most techniques. This leads to some confusion in this paper that should be addressed.
In this study, PM2.5 was analyzed at these two different pHs by two different instruments: pH 4.5, shown to be representative of European cloud water, and as a reference at pH 6.5 – this includes what is stated to be the optimal pH range for the ferrozine method. How does one interpret this measurement data since the pH of PM2.5 that entered the instruments would most likely be much lower than pH 4.5? Why were the measurements made at this pH, ie representative of cloud water? Is one to assume that the reported ws Fe concentrations represent what is in ambient PM2.5? It seems the assumption, (although not stated), is that no matter the pH or concentration of ws Fe in the ambient particles entering the instrument, the pH of the measurement determines the measured concentration of ws Fe. The logic is not clear, nor how one is to interpret soluble Fe concentrations in PM2.5 based on this data.
Another overall issue is the lack of clarity in terminology. A list of the terms with definitions in a table would be useful, and then take careful in how they are used throughout the paper.
Here is my understanding of the terms used.
Ws-Fe(II) is clear (line 67). It is the Fe measured by the ferrozine method for any sample pH for short reaction times.
Ws Fe is not clear, lines 63... suggest it is the Fe measured in an aqueous sample after liquid filtration also at any pH. But it is not specified how the Fe in the filtrate was detected.
Total ws-Fe (line 70, and Eq 4)) is the measurement of Fe(II) with the ferrozine method after treatment with HA that converts Fe(III) to Fe(II) at any pH. But the conversion rate depends on time, (line 255) so HA reaction time must also be known and remain constant. Also, likely some composition dependence.
Reducible Fe(III), due to the above limitation with HA conversion of Fe(III) to Fe(II) being operationally defined. Reducible Fe(III) refers to the conditions of these FIA systems. But what about this effect on Total ws-Fe?
Total Fe (line 63) is the total Fe measured in a particle, no method is specified, ie ICPMS, or XRF, ??
Fe(II), at times this term is also used, which is unclear. Is this the measurement of ws-Fe(II) but only for the standards?
Finally, how does the sampling/analysis system not alter the ratio of Fe(II)/Fe(III) from what it was in the ambient aerosol. Line 81 notes in filter sampling the redox state of ws Fe may change. If the proportion of Fe(II) and Fe(III) in the two online sampling systems is altered, what are the implications of this?
Specific Comments
Line 6 in Abstract not clear. If the PILS-FIA system at pH 4.5 represent pH of cloud water? What does the measurement for the MARS-FIA system at pH 6.5 represent?
What about the importance of oxalate or other organic Fe ligands? Presumably, the ferrozine forms a stronger complex than Fe-Ox, and so displaces it. So any Fe-Ox is measured as ws Fe(II)? Is this correct? Explain line 289.
Why is Fe(II) and Fe(III) vs Fe2+ and Fe3+ used at various times? It seems random, or is there meaning in this?
Figure 2 schematic could use some clarification, ie label the 6 port valve and 2 way pinch valve (I assume these are the black circles). It is not clear how the 6 port valve works. Eg, in the figure, what is happening? What do the dotted lines mean? How is the sample in the sample loop injected – no pumps (peristaltic channels) are shown – showing these might help?
Figure 3. Is the vertical axis label correct? Is it total ws-Fe, not ws-Fe?
In Figure 4, state in the Fig caption the pH of the filter extracts. Was the pH of the MARS and Filter extracts the same?
Lines 352 to 360. If the argument is that the MARS is higher than the filter due to all the various possible artifacts for the longer filter analysis time at the high pH of 6.5, why not test this with lower pH for both systems?
Figure 5 caption, state the sampling method used, MARS (a), PILS (b).
Regarding the last section of the paper comparing the two measurement periods and seasonal variation; there are other studies on this, for example, https://doi.org/10.1289/EHP2182. https://pubs.acs.org/doi/10.1021/acs.est.0c00483