the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Wall loss of semi-volatile organic compounds in a Teflon bag chamber for the temperature range of 262–298 K
Longkun He
Wenli Liu
Jixuan Wang
Mikinori Kuwata
Yingjun Liu
Abstract. Teflon bag chambers have long been used for investigating atmospheric chemical processes, including secondary organic aerosol formation. Wall-loss process of gas-phase species in Teflon bag chambers has typically been investigated at around room temperature. Recent laboratory studies started employing Teflon bag chambers at sub-273 K conditions for simulating wintertime and upper tropospheric environments. However, temperature dependence in vapor wall-loss processes of semi-volatile organic compounds (SVOCs) in a Teflon bag chamber has not well been investigated. In this study, we experimentally investigated wall-loss process of C14-C19 n-alkanes in a 1 m3 Teflon bag for the temperature range of 262 to 298 K. Enhanced wall losses of the tested n-alkanes were observed following the decrease in temperature. For instance, 65 % of C14 n-alkane was lost to the wall 15 hours after injection at room temperature, while the corresponding value was 95 % at 262 K. The experimental data were analyzed using the two-layer kinetic model, which considers both absorption of gas phase species to the surface layer of Teflon wall and diffusion to the inner layer. The experimental data demonstrated that absorption of gas phase species by the surface layer enhanced at lower temperature. The temperature dependence in absorption was well accounted using the equilibrium dissolution model of organic compounds to the Teflon surface by considering reduced saturation vapor pressure at lower temperature. On the contrary, diffusion process of n-alkanes from the surface to inner layer slowed down at reduced temperature. Hence the relative importance of the surface and inner layers on wall-loss process changes with temperature. Mechanistic studies on these processes will need to be conducted in the future to quantitatively predict the influence of temperature-dependent wall-loss processes of SVOCs on laboratory experimental results.
- Preprint
(1455 KB) - Metadata XML
-
Supplement
(901 KB) - BibTeX
- EndNote
Longkun He et al.
Status: open (until 12 Oct 2023)
-
RC1: 'Comment on amt-2023-187', Anonymous Referee #1, 18 Sep 2023
reply
General Comments
He et al. provide a good preprint in terms of scientific significance and presentation quality. However, I feel that detail is lacking, making scientific quality fair. I cannot currently recommend publication in Atmospheric Measurement Techniques until the specific points below are addressed, some of which I consider to be major revisions, though the editor may consider as minor.
Specific Comments
He et al. verify that a two layer model, as detailed in Huang et al. (2018) (https://doi.org/10.1021/acs.est.7b05575), with parameter fitting, can reproduce chamber observations of gas-wall partitioning under varying temperatures. Furthermore, they demonstrate that the absorption and diffusion mechanisms of wall loss of uncharged organic molecules have opposite responses to temperature change under dry conditions. The finding around diffusion response to temperature is novel, and it is in the interests of the chamber community that it is published.
Their observations that wall loss of semi-volatile organics is enhanced at lower temperatures is not, to my knowledge, novel, and has been reported by Zhang et al. (2015) (https://doi.org/10.5194/acp-15-4197-2015) (who identify a link with pure component saturation vapour pressure, but don’t conclude a direct causal relation with pure component saturation vapour pressure), and is observed in the supporting material of Huang et al. (2018) (https://doi.org/10.1021/acs.est.7b05575) (though they did not attribute the cause to changed pure component saturation vapour pressure). Furthermore, Matsunaga and Ziemann (2010) (https://doi.org/10.1080/02786826.2010.501044) show the relationship between fraction partitioned to wall and component saturation vapour pressure in their Figure 6, implying that whether the saturation vapour pressure changes because of a change of component or because of a change in temperature, the partitioning will change accordingly.
However, Huang et al. (2018) (https://doi.org/10.1021/acs.est.7b05575) state that the effect of temperature on gas-wall partitioning requires further study, and this paper is the first to my knowledge to fulfil this request and provide mechanistic insight. It is therefore a valuable paper.
I therefore recommend that the title be changed to indicate that mechanistic insight is presented. This way readers will be directed to this article when interested in the mechanisms of gas-wall partitioning in Teflon chambers.
I recommend that units in terms of minutes be converted to seconds (e.g. k1, k-1, k2), to be consistent with the literature (e.g. Huang et al. (2018) (https://doi.org/10.1021/acs.est.7b05575)).
I also recommend that the mechanistic aspect of greatest novelty – the diffusion through the Teflon interior – is expanded on in the paper. It seems quite feasible to plot k2 as a function of the effective diffusion coefficient of organics through the Teflon, along with a line/curve of best fit and its coefficients, as Huang et al. (2018) (https://doi.org/10.1021/acs.est.7b05575) do in their Figure 5. Additionally, then plotting diffusion coefficient (rather than k2 as currently given in Figure 7) against component saturation vapour pressure allows for a better mechanistic understanding and much easier comparison with other publications. The text of the ‘Results and discussion’ section should then be updated to reflect these new figures. If these changes around diffusivity cannot be implemented, the paper should explain why so that future research is able to improve on these experiments. In addition, if these changes around diffusivity cannot be implemented, then the abstract should be changed to discuss the Teflon inner layer interaction in broader terms, rather than inferring that diffusion is the key process in the interaction.
He et al. make no mention of the effect of ageing (in their case previous experiments involving injection of alkanes), but for a paper discussing wall losses, this should be at least discussed, if not quantified, as it is in Matsunaga and Ziemann (2010) and in Huang et al. (2018).
The area/volume ratio of a chamber is frequently described in other papers as a key determinant in gas-wall partitioning, therefore to aid readers in their interpretation, it would be useful to have this value given in units of /m in the section describing the chamber. I also find the explanation of enhanced partitioning compared to the Matsunaga results on line 175 to be too brief, and wonder whether the authors could either explain why chamber volume makes a difference, or consider an explanation in terms of area/volume rather than just volume.
Line 88 says that air leaked out of the bag. The authors must provide at least qualitative, and preferably quantitative, evidence that this did not significantly affect the concentration of alkanes in the bag, otherwise a leak of alkanes from the bag could cause the observed decreases in concentration with time, rather than gas-wall partitioning. On this point, have the authors considered that the removal of air from the bag for instrument sampling led to decreased pressure in the bag (rather than air leaks), which in turn led to compression of the bag volume? In this case, pressure inside the bag may have been maintained and therefore gas-phase concentration of alkanes were unaffected by changes in bag volume.
Line 66 says fans were used, and Figure 1 shows these fans to be outside the bag but inside the chamber housing. The authors must mention how the resulting buffeting of the Teflon bag affects the rate of mixing of air in the bag and therefore the rate of gas-wall partitioning, especially when comparing to other results, such as Matsunaga and Ziemann (2010).
Given that some of these recommendations are substantial, I expect that the authors will consider making appropriate changes to text throughout the paper beyond the specific locations of text mentioned here.
Technical Corrections
Line 12: consider ‘The wall-loss process’ rather than ‘Wall-loss process’
Line 23: consider ‘diffusion of n-alkanes’ rather than ‘diffusion process of n-alkanes’
Equation 7 (Line 131): should the Cgas term be present in the divisor on the left hand side?
Line 143: ‘fitted’ rather than ‘fited’
Line 228: consider ‘composed of FEP film’ rather than ‘composed of the FEP film’
Citation: https://doi.org/10.5194/amt-2023-187-RC1 -
RC2: 'Comment on amt-2023-187', Anonymous Referee #2, 25 Sep 2023
reply
This manuscript investigates the wall loss of organic gases in FEP Teflon chambers at temperatures below 298 K. This is a previously identified gap in our understanding of vapor wall loss in chamber simulations of atmospheric chemistry, and the importance of this question makes studies such as this one essential. The manuscript provides good, useful information about the effects of lower temperature on wall loss, but the study design leaves some significant gaps in understanding that could be improved with revision and potentially some additional experiments.
Critically, the authors first inject n-alkanes at room temperature and then spend several hours cooling the chamber. Given prior literature on the effect of partitioning to Teflon polymer at temperatures above 298 K, our knowledge of how diffusion coefficients of organic compounds in polymers vary with temperature, and the other conclusions drawn by the authors from this data, I am concerned that the partitioning observed at t = 3 hours may be highly path-dependent, and might differ from the partitioning that would occur if the alkanes had been injected into a pre-cooled chamber.
Investigation of gas-wall partitioning in Teflon tubing at 120 C (Mattila et al https://doi.org/10.1080/10962247.2023.2174612) has shown that partitioning delays did not vary with temperature. This indicated that the increase in vapor pressure of analytes was largely compensated by an increase in C_w (or C_FEP_Surface in the authors’ notation). This result suggests that at lower temperatures, one might observe that the decrease in vapor pressure would be offset by a *decrease* in C_w. However, the authors here observe consistent slopes in Figure 5, indicating that the C_FEP_surface/gamma ratio is remaining constant with temperature. And following the authors literature review, the lack of temperature dependence in gamma suggests that C_FEP_Surface is constant below 298 K.
The authors’ retrieval of k_2 (the first-order rate constant representing diffusion in to the bulk polymer) in Figure 7 indicates that at lower temperatures the rate of diffusion into the polymer is slowed. This is consistent with the results of Matilla et al: where at high temperatures there is more polymer available for partitioning, and then here at lower temperatures there is a restriction in the movement of analyte through the Teflon.
The potential mechanism that concerns me is that C_FEP_Surface may be lower at colder temperatures (less polymer available for sorption), but since the authors’ experimental design first establishes equilibrium at higher temperatures, there is ‘extra’ alkane locked up in the polymer at depths that would not be accessible within 3 h if the injection had been done in a pre-cooled freezer. Given the long timescales for evaporation from FEP, it seems plausible that as the chamber is cooled, the diffusion coefficients drop, and the surface layer potentially shrinks there could be significant mass transfer limitations keeping sorbed alkane at a given depth. In a two-layer model, this would be equivalent to transfer into the bulk polymer due to a shift in the dividing point between the two layers. This would significantly overstate the amount of wall loss compared to a case where the chamber was already cooled at the time an S/IVOC was introduced
To support the authors’ conclusion that C_FEP_Surface is constant with temperature, authors need to demonstrate that the amount of alkane sorbed in the walls at 3 hours is not a path-dependent process. Ideally this would be done through injection of analyte into a pre-cooled chamber. This could be done with just the most volatile alkanes to avoid any issues with nucleation. If this is not an option, another approach would be showing that the partitioning equilibrium at 3 hours is not dependent on cooling rate. In the methods, the authors wait an hour after injecting before cooling the freezer. Alternate approaches could be eliminating this one hour wait; and conversely slowing the cooling rate by gradually lowering the setpoint of the freezer. If the authors observe that the partitioning at 3 hours is consistent across these cases, it would be strong evidence in support of their result that C_FEP_Surface does not vary with temperature. If faster cooling (or pre-cooling) gives a significant decrease in the amount of alkane sorbed, that would indicate that there is a strong temperature dependence in C_w, consistent with prior literature.
Line 171: Prior literature has established that C_FEP_Surface does in fact scale directly with chamber surface area to volume (SAV) ratio. Authors should directly compare the SAV ratio of the Matsunaga & Ziemann chamber to their own, and report if the C_FEPSurface results are consistent
Line 207: How does the C_FEPSurface value compare to the Huang et al recommendation of C_w = 10.8 * A / V ?
Line 211: Literature values for gamma_inf are all calculated within Huang et al. 2018, who assume a fixed C_FEP_Surface (C_w). References and phrasing here should be updated to reflect the source of the calculated gamma_inf values and also mention the assumed C_w value.
Citation: https://doi.org/10.5194/amt-2023-187-RC2
Longkun He et al.
Longkun He et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 266 | 58 | 14 | 338 | 23 | 9 | 13 |
- HTML: 266
- PDF: 58
- XML: 14
- Total: 338
- Supplement: 23
- BibTeX: 9
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1