Abundances, emissions, and loss processes of the long-lived and potent greenhouse gas octaﬂuorooxolane (octaﬂuorotetrahydrofuran, c -C 4 F 8 O) in the atmosphere

. The ﬁrst observations of octaﬂuorooxolane (octaﬂuorotetrahydrofuran, c -C 4 F 8 O), a persistent greenhouse gas, in the atmosphere are reported. In addition, a complimentary ✿✿✿✿✿✿✿✿✿✿✿✿ complementary ✿ laboratory study of its most likely atmospheric loss processesand ✿ , ✿✿ its infrared absorption spectrum ✿ , and global warming potential (GWP) are reported. First atmospheric measurements of c -C 4 F 8 O are provided from the Cape Grim Air Archive (41 ◦ S, Tasmania, Australia, 1978–present), supplemented by two ﬁrn air samples from Antarctica, in situ measurements of ambient air at Aspendale, Victoria (38 ◦ S)

x 10 19 molecule cm -3 .Assuming no absorption from possible sample impurities, the c-C 4 F 8 O absorption cross section at 210 nm was found to be 6.25 x 10 -24 cm 2 molecule -1 .As discussed in Section 3.2, such a low cross section is susceptible to overestimation due to the presence of impurities in the c-C 4 F 8 O sample.Therefore, this spectrum was considered as an upper-limit in our lifetime analysis.
Reviewer Comment: P9L16 The problem with this measurements of the relative rate coefficient is that while measureable amounts (up to 10 %) of CHF3 are removed, the changes in the C4F8O concentration are too low to measure reliably.Could this have been improved by a better choice of reference compound (i.e. one that reacts more slowly with O(1D)?Why was the experiment stopped after only 10% of CHF3 was depleted ?Which absorption bands of C4F8O and CHF3 were used to derive the fractional losses?Unless there are good arguments against, the authors should consider doing further experiments to nail down this number.Alternatively, they might consider using the correlation between ionisation potential and O(1D) rate coefficient that is frequently used to estimate the latter.
Reply: Regarding the O( 1 D) reaction, presently there is not a more reliable reference compound to be used.Note that it is extremely difficult to accurately measure small O( 1 D) rate coefficients.It is also worth noting that an improved measure of the O( 1 D) reactive rate coefficient for the furan reaction would not significantly alter our conclusions regarding the atmospheric loss of the furan.In this work, a conservative upper-limit rate coefficient was reported that can be refined in the future, if warranted.Regarding the extent of reaction: The 248 nm photolysis of ozone was used as a source of O( 1 D) in the photochemical reactor.The introduction of ozone to the reactor was accompanied by an addition of molecular oxygen.After multiple ozone additions, the reaction of O( 1 D) by O 2 can dominate the O( 1 D) loss, making the O( 1 D) + furan and O( 1 D) + CHF 3 reactions less significant.

Regarding the absorption bands:
The absorption bands used in this study were 1120-1000 and 1180-1120 cm -1 for c-C 4 F 8 O and CHF 3 , respectively.Regarding further laboratory studies of the O( 1 D) reaction: Based on our experimental data and analysis, the O( 1 D) reaction will have a minor impact on the furan global atmospheric lifetime.Therefore, at our current level of understanding of the atmospheric chemistry and modeling (i.e., lifetime determination for persistent compounds) further refinement of the O( 1 D) rate coefficient is not warranted at this time.
The sentence "the precision of the infrared spectral subtractions are the primary sources of uncertainty in the measurements."will be replaced by "the precision of the infrared spectral subtractions are the primary sources of uncertainty in the measurements.The low conversion of the c-C 4 F 8 O and CHF 3 achieved in these experiments was primarily due to the build up of O 2 associated with the addition of ozone to the reactor, making the loss of O( 1 D) by reaction with O 2 more significant than its reaction with c-C 4 F 8 O and CHF 3 .The absorption bands used in this study were 1120-1000 and 1180-1120 cm -1 for c-C 4 F 8 O and CHF 3 , respectively."P9L16, at the end of the paragraph of the Section 3.3, the sentence "Given the small O( 1 D) rate coefficient for the reaction between O( 1 D) and c-C 4 F 8 O, a refinement of the measured O( 1 D) reactive rate coefficient will have a negligible impact to the total atmospheric loss of c-C 4 F 8 O." is suggested to be added.
Reviewer Comment: P9L24 The authors state that reaction with OH will not represent a loss of C4F8O in the atmosphere.I agree, but the authors should state why this is most likely to be the case.Will a perfluorinated furan react with OH like other fully fluorinated organics?What upper limit to the OH-rate coefficient would be necessary for OH reaction to compete with O(1D) induced losses?
Reply: As stated, the loss of c-C 4 F 8 O by reaction with OH radicals is most likely a negligible process.Due to laboratory challenges, it is not possible to measure values approaching 1 x 10 -17 cm 3 molecule -1 s -1 (equivalent to 3000 years lifetime).We estimate that our laboratory measurements would, at best, yield a rate constant upper-limit of ~1 x 10 -14 cm 3 molecule -1 s -1 , which would be a gross overestimate of the true rate constant.Reporting such a rate constant would be very misleading and possibly misinterpreted by the community.We did not attempt to conduct OH reactivity measurements of this compound for these reasons.The expected low-reactivity of the furan is supported by its low O(1D) reactivity.To compete with O(1D) induced losses, the OH rate coefficient would be equivalent to >1 x 10 -18 cm 3 molecule -1 s -1 .
Actions Taken: P9L22, the sentence "The loss of c-C 4 F 8 O via reaction with the OH radical is assumed to make a negligible contribution to the global lifetime and consequently we ignore the last term in Equation 4." is planned to be replaced by "The loss of c-C 4 F 8 O via the reaction with the OH radical is assumed to make a negligible contribution to the global lifetime in our analysis.the OH rate coefficient would need to be ~1 x 10 -17 cm 3 molecule -1 s -1 (equivalent to a 3000 year lifetime) to make a significant global lifetime contribution.Such a low rate coefficient represents a significant challenge to current rate coefficient measurement methods.Additionally, an expected OH low-reactivity of the furan is supported by its low O( 1 D) reactivity measured in this work.Additional laboratory studies that are beyond the scope of the present work would be needed to quantify the OH reaction." Reviewer Comment: P10L3 The authors state that is reasonable to assume a Lyman-alpha crosssection of about 1x 10-17 as it is "roughly consistent" with highly fluorinated compounds.As the authors go on to conclude that this process has the shortest associated lifetime, I find this unacceptably vague.What is the physical basis for assuming that a fluorinated furan will absorb at 121.6 nm with the same cross section as non-heterocyclic, perfluorinated gases?Measurement of the cross section at this wavelength is not impossible.The NOAA lab certainly has VUV capability (e.g. for 185 nm measurements using Hg-lines) which could be extended to 121.6 nm.Surely even a rough experiment is better than a precarious assumption.
Reply: The Lyman-alpha absorption cross section was not measured as part of the present study.
We are implying that it should be measured in future studies, which would refine our analysis.In the absence of experimental data, we surveyed available cross section data for highly fluorinated compounds and found that a cross section of 1 x 10 -17 cm 2 molecule -1 would represent a reasonable cross section upper-limit.Using this cross section value in our lifetime analysis provides a Lyman-alpha photolysis lifetime of ~4500 years.Assuming a smaller or greater cross section would yield slightly longer or shorter lifetimes, although the relationship is not linear due to the dependence on the transport time to the mesosphere.Therefore, in the absence of experimental data, we believe our assumed cross section and lifetime is reasonable.Laboratory studies to measure VUV cross sections would help refine our analysis but not alter the primary conclusion from this work that the atmospheric lifetime of the furan is extremely long.Note that measuring VUV spectra in the laboratory is significantly different than measuring 185 nm cross sections and requires specialized equipment.
Actions Taken: P10L2, the sentences "It is reasonable to assume a Lyman-α cross section of 10 −17 cm 2 molecule −1 for c-C 4 F 8 O which would be roughly consistent with values for highly fluorinated compounds (SPARC, 2013).The estimated lifetime due to Lyman-α photolysis, τ Lyman−α , is then ~ 4500 years (a smaller Lyman-α cross section would lead to a longer lifetime)."will be replaced by "The scope of the present study did not include a measurement of the Lyman-α cross section.It is however reasonable to assume a Lyman-α cross section of 10 −17 cm 2 molecule −1 for c-C 4 F 8 O which is in the range of values for highly fluorinated compounds, (0.035 -10) x 10 −17 cm 2 molecule −1 (SPARC, 2013).Therefore, in the absence of experimental data, we consider the Lyman-α cross section, 1x10 -17 cm 2 molecule -1 , used in our lifetime analysis as a reasonable estimate.Note that a smaller (larger) Lyman-α cross section would lead to a longer (shorter) photolysis lifetime, although the lifetime dependence on the cross section value is not linear due to the lifetime dependence on the transport time to the mesosphere." Replies to Review #2 for Interactive comment on "Abundances, emissions, and loss processes of the long-lived and potent greenhouse gas octafluorooxolane (octafluorotetrahydrofuran, c-C 4 F 8 O) in the atmosphere" by Martin K. Vollmer et al. By Anonymous Referee #2 (Received and published: 25 October 2018) Replies to the reviewer comments are added in blue color following each comment, and the revised text is in green.We thank the reviewer for his/her input.We believe that with the suggested changes to these valuable comments, the manuscript will improve.

Reviewer Comment: General comments:
The article presents a large dataset and budget estimate for a newly detected compound in the atmosphere: c-C 4 F 8 O.Although its abundance is small (less than 0.1ppt) its radiative efficiency is strong and lifetime likely very high.It is still unregulated and sometimes viewed as a promising compound in terms of industrial applications (see for example Kočišek et al., 2018).I think that its scope and novelty make it adequate for a publication in ACP.I have some comments on the methodology and presentation.
Reviewer Comment: The Northern Hemisphere (NH) measurements are little described and commented (p 1 l5-6, p3 l14-15).The article should explain how the North Hemisphere trend (dashed line on Figure 1) was constrained and evaluate the uncertainty on emissions resulting from the lack of NH constraints.
Reply: Northern Hemisphere samples are planned to be better described in the Supplement as follows: "S-2.5 Northern Hemisphere (NH) samples The Southern Hemisphere Cape Grim Air Archive (CGAA) samples were complemented with a few archived air samples from the Northern Hemisphere (see also Table S3).Some of these samples were collected as whole air ambient background samples for original calibration purposes: UAN920470 at Cape Meares, Oregon, most likely cryogenic techniques; T-EMPA-1 and J-187 at La Jolla, California using an oil-free diving compressor (Rix Industries); EG-003 at Jungfraujoch, Switzerland, using cryogenic techniques.H-160 at Mace Head, Ireland, using an oil-free diving compressor.These samples were all collected in internally electropolished stainless steel canisters (Essex Industries, USA).Two samples collected at Dubendorf (DUE161216-D2 and DUE161216) were collected into 6-L internally electropolished cylindrical custom-fabricated containers (LabCommerce, California) using a diaphragm pump (KNF-N-022-ANE, Neuberger), for the specific purpose of this project.These two samples, and EG-003 and H-160 were shipped from Empa to CSIRO for analysis along with the CGAA samples under same measurement conditions." We have now also better described how the NH trend was constrained.The text added is: "The c-C 4 F 8 O measurements in the Southern Hemisphere provide a strong constraint on the trend in both hemispheres due to the very long lifetime of c-C 4 F 8 O in the atmosphere, the relatively rapid mixing of the atmosphere, and the expectation that most c-C 4 F 8 O emissions are in the Northern Hemisphere (NH).Most anthropogenic gases are released predominantly in the NH, including gases released by the semiconductor industry.The assumption of mainly NH emissions for c-C 4 F 8 O leads to higher mole fraction values in the NH, and this is confirmed by comparison of the modelled NH history with the few NH measurements that we do have.In the Supplement, we test the sensitivity of inferred global emissions and mole fraction in both hemispheres to the assumed spatial distribution of emissions, and show the uncertainty in inferred mole fraction in both hemispheres."Reviewer Comment: Similarly, the mixing ratio and emission trends between 1950 and 1978 are mostly constrained by a single firn air data point undergoing a large age distribution, and having a mixing ratio (6 ppq, Table S4) very close to the detection limit (5 ppq, p4 l13).The article should explicitly discuss the constraints on the anthropogenic versus natural sources of c-C 4 F 8 O, as well as the little constrained early emissions.
Reply: We agree, and suggest to change the text in the results section to address these concerns: "This suggests that c-C 4 F 8 O was below 10 ppq in the Southern Hemisphere before 1978.However it is impossible to further pin down the first appearance of this compound in the atmosphere and the exact course of the abundance until ~1980 because our knowledge of c-C 4 F 8 O prior to the CGAA is based on only one firn sample measurement with air spanning several decades (see calculated Green's functions in the Supplement).Also, potential small contamination during firn air sampling by modern air or sampling devices cannot be fully excluded.Additionally, the measurement of the older firn air sample is close to the instrument's detection limit.Given these limitations, we are not able to draw any conclusions on any potential naturally-occurring c-C 4 F 8 O. Nevertheless, the two firn air sample measurements allow us to draw conclusions on storage stability of c-C 4 F 8 O in canisters.Given that the storage time of the two firn air samples in the canisters is much shorter than those of the older CGAA samples, the good agreement of the firn air results with those of the CGAA is supportive of storage stability of c-C 4 F 8 O in the CGAA tanks and confirms that the observed multidecadal record is not a simple artifact of degradation of c-C 4 F 8 O in canisters over time." In addition we suggest to show the early history with dotted lines rather than solid lines,to emphasize the greater uncertainty before 1978.See revised figure further below.
Reviewer Comment: However, for a well-mixed very long lived species, a reasonable estimate of global emissions can be obtained from a simple one box model calculation.Presenting this simple calculation and comparing it to the elaborate approach used would improve the description of the main uncertainties and be helpful to non-specialist readers.
Reply: We believe that adding another model calculation would be rather confusing than helping to understand the main uncertainties.The 12-box model has been used and validated many times in the past in numerous publications and doesn't need to be re-assessed here.Nevertheless, we have made a quick intercomparison based on an 1-box model using the following assumption: 1.8E20 mol of air in the total atmosphere; well mixed (no delay in stratosphere); no sinks; using the fitted observations from the 12-box model.Admittedly, the latter assumption creates some degree of dependency to the 12-box model, but an independent fit through the observations would not significantly alter the results.The result is shown here in the subfigure c) as orange line.Some of the discrepancies to the emissions from the 12-box model is likely caused by the above assumption, in particular that of a uniform vertical atmosphere.In addition, we have calculated the cumulative emission in a 1-box model approach using the end-of record mole fractions of about 74.5 ppq.Again using 1.8 E20 mol of air in the total atmosphere and no sinks for c-C 4 F 8 O, we calculate 2.89 kt, which compares well with the 2.85 kt from the 12-box model.We propose to not mention these 1-box model results in the revised text for the reasons mentioned above.
Reviewer Comment: A first estimate of the lifetime of c-C 4 F 8 O is provided but some important assumptions should be better described: the basis for the estimated Lyman-α lifetime and OH reactivity (comparison with species having similar bonding structures?), the possible role of other unexplored sinks such as surface loss (to ocean and land) and heterogeneous processes should be discussed at least in terms of perspectives.
Reply: Thank you for addressing these points.We handle the comments regarding Lyman-alpha and OH reactivity as part of the replies to reviewer 1, and would like to refer to that reply.With regard to other possible sinks, we suggest to add the following sentence to the Introduction (following the discussion on the lack of existing information on atmospheric loss of the compound): "Information is also lacking on other potential loss processes for c-C 4 F 8 O, such as uptake by oceans and land".
We also suggest to add the following sentence to the description of the 12-box model: "Loss processes other than those in the atmosphere, such as uptake by land and oceans, and potential natural sources, are not included in the model." Further, we suggest to add the following sentences to the discussion of the 'Atmospheric Lifetime (3.4)': "This study has focused primarily on the atmospheric loss processes of c-C 4 F 8 O, i.e., potential deposition or heterogeneous loss processes of c-C 4 F 8 O were beyond the scope of this study.Deposition or heterogeneous loss processes, if significant, would lead to a shorter global lifetime for c-C 4 F 8 O." Ultimately we modified the last sentence in the conclusions to: "However, even if emissions were completely halted, it will, under the assumption of insignificant non-atmospheric sinks, take thousands of years for c-C 4 F 8 O to be removed from the atmosphere." Reviewer Comment: Specific comments: p2 l10-11: The Californian regulation could be mentioned (https://ww2.arb.ca.gov/resources/documents/semiconductor-regulation)Reply: We have explored this a bit more.Rather than mentioning the Californian regulations, we have mentioned the US EPA regulations and IPCC.The revised text is suggested as "The compound is listed in the Intergovernmental Panel on Climate Change (IPCC) 2006 guidelines in support of UNFCCC (IPCC, 2006, Volume 1, Chapter 8) as a compound, for which countries are encouraged to provide emissions estimates (on a mass unit until a published greenhouse warming potential (GWP) will become available).In the 2013 Revisions of the UNFCCC reporting guidelines (UNFCCC 2013), c-C 4 F 8 O is absent from the list of compounds with mandatory reporting.Additional reporting regulations exist on country or state levels.In the USA, large suppliers and emitters of c-C 4 F 8 O are required to report the amounts they supply or emit under the Greenhouse Gas Reporting Program (GHGRP, URL: https://www.epa.gov/ghgreporting,accessed January 2019).When CO2-equivalent emissions are required for these submissions, a default GWP for fully fluorinated GHGs of 10,000 is used due to the lack of a peer-reviewed GWP.Emissions have mainly been reported under the ``Fluorinated Gas Production'' subpart for 2011--2017 with a maximum of 40 t in 2013 and a subsequent decline to 4.5 t by 2017" Reviewer Comment: p5 l11-14: As pumping out the interstitial air from deep firn can be difficult and induce contamination, more indications should be provided about the multi-species consistency of model results for the deep firn air sample used and the overall firn.For example, the RMSD/σ indicator used in Buizert et al. ( 2012) could be provided.The reason why so few depth levels were analyzed for c-C 4 F 8 O should be given, sample size issue?
Reply: The two ABN firn measurements play only a small role in this study.It is discussed above that the early history is not well constrained by the single deep firn measurement due to its age spread, and that contamination cannot be ruled out.Another publication is underway that will describe the ABN measurements and modelling in much greater detail, including showing how well the firn model fits all measurements used for calibration.The level of detail suggested by the reviewer is not seen as necessary for this study given the small role of the firn measurements, so no further change has been made.
We suggest to add the following sentence to explain why there were only 2 samples available for this project (in Methods)."Only two samples were available for the present study as other samples from this site were used for a different halocarbon study." Reviewer Comment p5 l14-18: The Trudinger et al. (2013) model uses both molecular and eddy diffusivity terms.As this has the same effect as modifying the diffusion coefficient, the relative roles of molecular and eddy diffusivity terms for the ABN firn should be commented.
Reply: We plan to comment on this by adding the following: "Only molecular diffusion was used for the ABN firn model calculations; eddy diffusivity is sometimes used in the deep firn but was not used here as the parameters were not well constrained by the available measurements".
Reviewer Comment: p5 l19-26: how were the North Hemisphere concentrations evaluated?Reply: We are not clear about this comment.We assume that it is related to the first comment about the NH concentrations, and believe to have addressed this comment sufficiently there.
Reviewer Comment: p5 l29: Vollmer et al. (2016Vollmer et al. ( , 2018) ) used multi-depths firn air constraints from both hemispheres.The methodological adaptations to the lack of NH constraints should be described.
Reply: The method was not adapted from Vollmer et al (2016Vollmer et al ( , 2018) ) due to the use of only SH constraints.As described above, the NH trends are well constrained by SH measurements due to the long lifetime and predominantly NH emissions.
Reviewer Comment: p6 l1: I do not understand what the Green's functions from the 12-box model are and did not see an explanation in Vollmer et al. (2016Vollmer et al. ( , 2018) ) Reply: To clarify this, we suggest to add the following text "Green's functions derived from the 12box atmospheric model relate atmospheric mole fraction in the high-latitude Northern and Southern Hemispheres to annual global emissions in preceding years, and are used in the inversion (Trudinger et al;, 2016)." Reviewer Comment: p6 l11-13: The emission values in Ivy et al. (2012) start in 1980 (Table 3), how was the prior estimate designed for the 1950-1980 period and what impact does it have on the final solution for this weakly constrained period?
Reply: We suggest to add some text to correct for this shortfall.Although of importance, we consider it a detail that better fits into the supplement, in particular in relation to (original) Fig. S4, where we graphically show what we did.The suggested text is: "We construct a c-C 4 F 8 O prior history from emissions of perfluorooctane because this compound has similarly low abundances and a long lifetime as c-C 4 F 8 O.For our standard case, we use perfluorooctane emissions published by Ivy et al. (2012) for the 1980 -2010 period with the perfluorooctane 2010 value as a constant value for 2010 -2017 and the 1980 value for perfluorooctane for the 1950 -1980 period.We also test the sensitivity of our results to a number of other prior histories.a) the standard case doubled, b) the standard case halved, c) the standard case with emissions before 1980 extrapolated back to zero in 1950 and d) a small linearly increasing function (all shown in Fig. S4).Our analysis shows that the emissions derived for c-C 4 F 8 O are rather insensitive to the choice of the prior, because the prior is used as a starting point for the inversion only, and not as a constraint." Reviewer Comment: p7 l23 and after, including section 2.2 of the Supplement: a single notation should be adopted to name reaction rates, avoid using kR, then k1 (implicit) and k2, then kc-C 4 F 8 O.
Reply: We agree and suggest the following revision: "kR" will be replaced by "k 1 ", and "k c-C4F8O " and "k CHF3 " will be replaced by k 1a and k 2a , respectively in the Equation II.In the supplement, "k/k CHF3 " in Table S5 will be replaced by "k 1a /k 2a ".In the footnote of Table S5 Reviewer Comment: p10 l9: Figure 4 is little commented, it could be shifted to the Supplement or combined with Fig. 2 Reply: We prefer the Figure 4 to remain in the manuscript.It shows how IR absorption bands fall within the atmospheric window qualitatively and quantitatively.It is also used as a basis for Radiative Efficiency calculations and the discussions on GWP (p. 10 L. 9-13).
Reviewer Comment: p10 l26-27: circular argument, the calculated growth rate is small because the measured concentration trend is weak (in recent years), not the contrary.
Reply: Thank you for spotting this.We suggest to change the sentence(s) to: "The growth rate was at a maximum of 4.3 ppq/yr in 2004 and declined from that to <0.15 ppq/yr in 2017 as a consequence of the relatively constant abundances in the last few years." Reviewer Comment: p11 l6-9: the important Aspendale dataset (thousands of measurements) is briefly summarized in Table S3 and very briefly commented.A more in-depth discussion of c-C 4 F 8 O variability at various sub-annual time scales and recent trend, as well as a plot (at least in the Supplement) would be useful.
Reply: We plan to address this comment by providing an additional section in the supplement including a figure showing the high-resolution data set from in-situ measurements at Aspendale.There is no in-depth discussion on sub-annual time scales and trends, as there is no variability for this record, which we have already stated in the main text.
The revised text and figure is suggested as follows: "S-2.6 In-situ measurements of c-C 4 F 8 O at Aspendale Regular measurements of c-C 4 F 8 O in ambient air at Aspendale were started in February 2017.These were conducted on a 2-hourly basis whereas each air measurement is bracketed by standard measurements.Results are shown in Fig. S4.A few ambient air measurements were also made in late 2016 during the CGAA measurement phase.These were made from 3 L samples (vs the regular 2 L samples) and show improved precisions compared to the remaining record.The 2year record shows constant c-C 4 F 8 O mole fractions within the precisions of the measurements.There is no sign of any pollution events in this record suggesting that there are no significant sources of c-C 4 F 8 O within the footprint of the site.Furthermore and given the long atmospheric lifetime of the compound, the absence of a significant trend is suggestive of the absence of major global emissions in the last years."Reviewer Comment: p11 l26: the wording "a few other synthetic greenhouse gases" implicitly assumes that c-C 4 F 8 O is purely anthropogenic but this is not discussed in the article Reply: The reviewer is correct.We suggest to remove the word 'synthetic'.
Reviewer Comment: p12 l6: due to the high cost of Antarctic field operations, research programs and logistic institutions financing them are usually explicitly named.
Reply: While we agree with the reviewer on the large-scale operations of the Antarctic field programs, this has to be also viewed in ratio to other contributions and be somewhat balanced.For example, compared to the two firn air samples, the input from the general AGAGE operation is large, AGAGE is a very large and costly long-term network and yet we cannot acknowledge all of its sub-contributions and funding agencies.(parts per billion, nanomol mol −1 ) was determined, which is one of the largest found for synthetic greenhouse gases.The global annually averaged atmospheric lifetime, including mesospheric loss, is estimated to be >3 000 years.GWPs of 8 975, 12 000, and 16 000 are estimated for the 20, 100, and 500-year time-horizons, respectively.
The above applications have emerged only within the last two decades.Whether c-C 4 F 8 O was used earlier than that is undocumented.Frick and Anderson (1972)  The present study aims to improve our knowledge on the chemical and radiative properties of c-C 4 F 8 O relevant to determining its atmospheric lifetime and to provide the first atmospheric measurements from which we derive estimated global emissions to the atmosphere.Measurements were made on atmospheric samples archived in canisters and Antarctic firn, and in modern air from in situ observations.From the derived historical record, emissions are estimated using a 12-box chemical transport model of the atmosphere (Cunnold et al., 1983;Rigby et al., 2013;Vollmer et al., 2016).We also conducted labora- For the present study, archived and urban ambient air samples were analyzed at the Commonwealth Scientific and Industrial Research Organization (CSIRO) laboratory at Aspendale (Victoria, Australia) using Medusa gas chromatographic (GC) mass spectrometric (MS) techniques (Miller et al., 2008).The archived air samples consisted primarily of the Cape Grim Air Archive (CGAA) samples collected under clean air baseline conditions for archival purposes since 1978 at the Cape Grim Baseline Air Pollution Station (Tasmania, Australia 40.7 • S, 144.7 • E).These >100 samples were collected into 34 L internally electropolished stainless steel canisters (Essex Industries, USA) using cryogenic techniques (Fraser et al., 1991;Langenfelds et al., 1996Langenfelds et al., , 2014;;Fraser et al., 2016).The CGAA record was complemented with a few samples collected in the Northern Hemisphere In situ measurements of c-C 4 F 8 O at Aspendale (38.0 • S, 145.1 • E) were started in February 2017.These samples are collected from the rooftop at CSIRO (at 11 m height from the ground) through a 3/8" OD Synflex 1300 tube (Saint-Gobain, France) using a continuous flow air sampling module (Miller et al., 2008) with a diaphragm sampling pump fitted with stainless steel heads and a neoprene membrane (KNF Neuberger, Germany).
All archived air samples were analyzed on the Medusa-GCMS "Medusa-9" in December 2016.

✿
The instrument is based on the original design of the Medusa-GCMS used in the Advanced Global Atmospheric Gases Experiment (AGAGE) network (Miller et al., 2008;Prinn et al., 2018) but fitted with different chromatography columns (Vollmer et al., 2018).A GS-GasPro main capillary column (0.32 mm ID × 60 m, Agilent Technologies) was used for the main separation and a column of the same type (5 m) was fitted as a precolumn, allowing for a backflushing of late eluting compounds.In this GCMS setup (Agilent 6890 GC, 5975 MS) c-C 4 F 8 O was identified using a multi-component diluted mixture of known composition with the MS in scan and selected ion modes.The choice for the two fragments used in the analysis of our air samples was based on the mass spectrum, which we measured for c-C 4 F 8 O, to the best of our knowledge the first one published for this compound (see Supplement).
Analytes from the samples were cryogenically preconcentrated on a first microtrap of the GCMS and subsequently transferred to a second microtrap, both filled with HayeSepD and held at ∼ −155 • C.During this process, water vapor was largely removed using nafion dryers; nitrogen, oxygen, and a large fraction of noble gases were removed due to their trap breakthroughs, and carbon dioxide was removed using a molecular sieve (4A) packed column between the traps.To enhance the signal size of the measured compounds, 3 L sample sizes were used for each measurement (compared to normally 2 L) and the MS electron multiplier voltage was increased by 50 V compared to what was given by the autotune algorithm.Analysis of a single sample lasted 65 min.Archived air sample measurements were bracketed by measurements of a standard (E-146S) to track and correct for MS sensitivity changes.This standard was air compressed into a 34 L tank at the remote Rigi-Seebodenalp station (Switzerland) using an oil-free compressor, ✿ and was additionally spiked with small amounts of c-C 4 F 8 O and other compounds, ✿ to enhance the GCMS peak size and signal-to-noise ratio.In general, three measurements of each archived air sample were made.For some, no standard measurement was made between the second and third sample to assess potential memory effects of the system.For c-C 4 F 8 O, no memory effect and no signal in the blank runs could be detected.Detection limits are estimated at 5 ppq (parts per quadrillion, femtomol mol −1 ).Mean precisions (2σ) for the measurements of the archived air samples ranged 3-4 ppq (20-5 %) for the low (∼15 ppq) to high (∼70 ppq) mole fractions, respectively.Based on two different types of experiments, a linear system response for the relevant mole fraction range was found (see Supplement).In situ urban air measurements at Aspendale are based on 2 L samples and without alteration of the MS electron multiplier voltage.
Consequently the precisions are slightly poorer for these measurements.These air precisions were estimated at ∼12 ppq (∼17 %, 2σ) under the assumption that c-C 4 F 8 O remains constant in the air measured in situ at Aspendale on a daily basis.
A primary calibration scale was prepared based on a commercially obtained multi-component mixture in dry synthetic air (Carbagas, Switzerland, HCP-04Carba), with a mole fraction of c-C 4 F 8 O at 10 ppm (parts-per-million, µmol mol −1 ).This mixture was diluted manometrically and using a bootstrap technique, resulting in a primary calibration standard (EP-001) with c-C 4 F 8 O at 1.81 ppt (parts-per-trillion, picomol mol −1 ).Three secondary standards were additionally prepared from ambient air compressed into cylinders (Essex Industries, USA) and spiked with small quantities of c-C 4 F 8 O resulting in mole fractions of ∼0.5 ppt.These secondary standards were the base for propagating the calibration scale to other calibration standards, in particular that used for the Cape Grim Air Archive measurements (E-146S).They define the Empa-2013 calibration scale for c-C 4 F 8 O on which our results are reported.The systematic uncertainty of the preparation of this primary calibration scale (including its propagation to the working standards), which defines its accuracy, is estimated at 15 % (2σ).Details of the dilution technique and the primary calibration scale are provided by Vollmer et al. (2015).

Firn model
We use a numerical firn air model (Trudinger et al., 1997(Trudinger et al., , 2013) ) (2006)) and a multiplier for the Le Bas increments of 0.97 (this value minimizes the difference of calculated relative diffusion coefficients of a number of compounds from values measured by Matsunaga et al. (1993Matsunaga et al. ( , 2002Matsunaga et al. ( , 2005))).We use the AGAGE 12-box atmospheric model (Rigby et al., 2013) to relate the atmospheric mole fractions to surface emissions.Briefly, in this model, the atmosphere is divided into four zonal bands, separated at the equator and at the 30 • latitudes, thereby creating boxes of similar air masses.There are also vertical separations, at altitudes represented by 500 hPa and 200 hPa, resulting in the overall 12 boxes.Model transport parameters and stratospheric photolytic loss vary seasonally and repeat interannually (Rigby et al., 2013).We anticipate that variations in emissions dominate atmospheric trends, particularly over the longer (multi-annual) timescales that are our primary focus, so inter-annual variation in transport is not expected to be important here.

Global inversions
To estimate global emissions to the atmosphere from the mole fraction measurements, we employ an inverse calculation (inversion InvE2 from Trudinger et al. (2016), and termed "CSIRO" inversion in Vollmer et al. (2016Vollmer et al. ( , 2018))) that was developed to focus on sparse observations from air archives, and firn air and ice core samples that are associated with age spectra.The weighted by the observation uncertainties, plus the sum of the year-to-year changes in emissions (Trudinger et al., 2016).
Given the lack of industry-based bottom-up emission estimates for c-C 4 F 8 O, we use emissions derived from observations , of perfluorooctane, which was found present for many decades and at low abundances in the global atmosphere (Ivy et al., 2012).
Because the prior is not based on information on c-C 4 F 8 O, we do not include the prior in the cost function.The emissions derived from the inversion are rather insensitive to the choice of the prior (see Supplement), because the prior is used here as a starting point for the inversion only, and not as a constraint.Our observations used in the inversion are the firn measurements and annual values of mole fraction from a smoothing spline fit (50 % attenuation at 10 years) to measurements of the CGAA and in situ measurements at Aspendale.Northern Hemisphere measurements were compared with the reconstructed mole fractions for that hemisphere, but were not used in the inversion.Uncertainties for the CGAA and Aspendale annual means Infrared absorption spectra were measured at 296 K using Fourier transform infrared ✿✿✿✿✿✿✿ infraRed (FTIR) spectroscopy over the 500-4 000 cm −1 spectral region at 1 cm −1 resolution with Boxcar apodization.The apparatus has been used extensively in previous studies (Bernard et al., 2017(Bernard et al., , 2018a)).The FTIR was coupled to a 15 cm path length single pass absorption cell with potassium bromide (KBr) windows.A liquid-nitrogen cooled HgCdTe/B semiconductor detector was used.Infrared spectra were recorded in 100 or 500 co-added scans.Absorption spectra were recorded under static conditions using a dilute mixture
Provided c-C 4 F 8 O and the reference compound are removed solely by reaction with O( 1 D), the rate coefficient for reaction 1a is related to the reference compound rate coefficient by the equation The yield of the O( 1 D) channel is 0.9 (Burkholder et al., 2015).After thoroughly mixing the gas mixture in the system, a time zero infrared spectrum was recorded.Ozone was then added slowly to the reactor with the photolysis laser and gas circulation on.The photolysis laser fluence was in the range ∼2-7.4 mJ cm −2 pulse −1 .The laser was operated at 10 or 20 Hz.
The total pressure in the cell increased during an experiment by ∼300 Torr, mostly due to the addition of He carrier gas used to flush ozone into the reactor.Infrared spectra were recorded at regular intervals with approximately 10 spectra recorded over the course of an experiment.Experiments performed separately demonstrated that there was no significant loss of c-C 4 F 8 O or were in the range 6.4-6.8×10 14molecule cm −3 and 4.5-5.0×10 14molecule cm −3 , respectively.
3 Results and discussion

Infrared spectrum
The infrared absorption spectrum of c-C 4 F 8 O obtained in this study is shown in Fig. 1.Over the range of c-C 4 F 8 O concentrations used, the spectra obeyed Beer's law with high precision (∼0.2 %).Spectra recorded at different total pressures had identical band shapes, i.e., the spectrum was independent of the total pressure (He bath gas) over the range of 30-400 Torr.
The integrated band strength (IBS) over the spectral region 500-1 500 cm −1 was determined to be (3.21 ± 0.01) × 10 −16 cm 2 molecule −1 cm −1 , where the quoted uncertainty is the precision of the linear least-squares fit of the data to Beer's law (Equation I).The absolute uncertainty in the c-C 4 F 8 O spectrum includes estimated uncertainties in the optical path length (±0.5 %), measured absorbance (±0.005), temperature (±1 K), and pressure (±0.2 %).The absolute uncertainty in the total integrated band strength is estimated to be 3 %.

UV absorption
UV absorption of c-C 4 F 8 O was observed between 200 nm and 225 nm, a range that is most critical for calculations of the c-C 4 F 8 O atmospheric photolysis rates.The spectrum is continuous, with cross section ✿✿✿✿✿✿✿ sections decreasing monotonically with increasing wavelength.The cross section measurements obeyed Beer's law with values of (9.2 ± 3.9) × 10 −24 and (4.4 ± 2.3) × 10 −24 cm 2 molecule −1 cm −1 at 200 and 225 nm, respectively.✿ Overall, the cross sections of c-C 4 F 8 O were very low, and therefore, the measurements are susceptible to interference from even minor sample impurities.Therefore, we choose to assign a conservative UV cross section of <2 × 10 −23 cm 2 molecule −1 cm −1 over the 200-225 nm range.

O( 1 D) reaction
We found the reactivity of c-C 4 F 8 O with O( 1 D) to be low, which makes the determination of an accurate rate coefficient more challenging.The relative rate data are shown in Fig. 3 and tabulated in the Supplement.The precision of the three independent measurements is high with a fit precision of a few percent.However the agreement between the independent measurements is relatively poor.The low conversion of c-C 4 F 8 O, <2 %, and the precision of the infrared spectral subtractions are the primary sources of uncertainty in the measurements.The

Atmospheric lifetime
The global annually averaged atmospheric lifetime (τ ) of c-C 4 F 8 O, is defined with respect to the individual partial lifetimes by the relationship:  (Myhre et al., 2013).
The GWP of c-C 4 F 8 O was calculated using the global atmospheric lifetime lower-limit of 3 000 years and the radiative efficiency determined in this work: where RE is the radiative efficiency, T is the time horizon (in years), M  as derived below, they correspond to 34 Mt CO 2 -equivalents.Despite the high GWP, these emissions are small compared to the major greenhouse gases but of similar magnitude to some of the other minor greenhouse gases such as minor perfluorocarbons and fluorinated inhalation anesthetics (Ivy et al., 2012;Vollmer et al., 2015).Whether these cumulative emissions remain at low levels will depend on potential future choices for c-C 4 F 8 O in large scale applications.

Figure S2 .
Figure S2.Measured 296 K gas-phase UV spectrum of the octafluorotetrahydrofuran (c-C 4 F 8 O) sample used in this study.The c-C 4 F 8 O concentration was 2.16x 10 19 molecule cm -3 .Assuming no absorption from possible sample impurities, the c-C 4 F 8 O absorption cross section at 210 nm was found to be 6.25 x 10 -24 cm 2 molecule -1 .As discussed in Section 3.2, such a low cross section is susceptible to overestimation due to the presence of impurities in the c-C 4 F 8 O sample.Therefore, this spectrum was considered as an upper-limit in our lifetime analysis.

Figure S4 .
Figure S4.Ambient air measurements of c-C4F8O at Aspendale (Victoria, Australia, 38.0 °S, 145.1 °E).The measurements are expressed as dry air mole fraction in parts-per-trillion on the Empa-2013 calibration scale.Results show constant c-C 4 F 8 O within the precision of the measurement.

✿
Correspondence to: Martin K.Vollmer (martin.vollmer@empa.ch)Abstract.The first observations of octafluorooxolane (octafluorotetrahydrofuran, c-C 4 F 8 O), a persistent greenhouse gas, in the atmosphere are reported.In addition, a complimentary warming potential (GWP) are reported.First atmospheric measurements of c-C 4 F 8 O are provided from the Cape Grim Air Archive (41 • S, Tasmania, Australia, 1978-present), supplemented by two firn air samples from Antarctica, in situ measurements of ambient air at Aspendale, Victoria (38 • S), and a few archived air samples from the Northern Hemisphere.Atmospheric abundances ✿✿✿ The (parts per quadrillion, femtomol mol −1 in dry air) by 2017.However its growth rate ✿✿✿✿✿✿✿✿✿✿ 2015-2018.✿✿✿✿ The ✿✿✿✿✿✿ growth ✿✿✿ rate ✿✿✿ of c-C 4 F 8 O has decreased from a maximum in 2004 of 4.3 ✿✿✿ 4.0 ✿ ppq yr −1 to <0.15 ✿✿✿✿ 0.25 ppq yr −1 in 2017.Using a 12-box atmospheric transport model, globally averaged yearly emissions and abundances of c-C 4 F 8 O are calculated for 1951-2017 ✿✿✿✿✿✿✿✿✿ 1951-2018.Emissions, which we speculate to derive predominantly from usage of c-C 4 F 8 O as a solvent in the semiconductor industry, peaked at 0yr −1 in 2004 and have after declined to <0.01 ✿✿✿✿ kt, which correspond to 34 Mt of CO 2 -equivalent emissions.Infrared and ultraviolet absorption spectra for c-C 4 F 8 O as well as the reactive channel rate coefficient for the O( 1 D) + c-C 4 F 8 O reaction were determined from laboratory studies.On the basis of these experiments, a radiative efficiency of 0.430 W m −2 ppb −1

✿✿✿✿✿(
SF 6 ), and nitrogen trifluoride (NF 3 ) are important anthropogenic greenhouse gases, which are included in the Kyoto Protocol to the United Nations' Framework Convention on Climate Change (UNFCCC).Although they don't ✿✿ do ✿✿✿ not ✿ have the capacity to destroy stratospheric ozone (unlike e.g.chlorofluorocarbons), HFCs have also been added to the Montreal Protocol on Substances That Deplete the Ozone Layer through the recent Kigali Amendment 4 F 8 O has been under discussion in the recent literature foremost as a new Chemical Vapor Deposition (CVD) chamber patented a method to synthesize c-C 4 F 8 O for potential use as an inert solvent 2 for highly reactive or corrosive halogenated materials in naval applications.However it remains unclear if this led to mass production of c-C 4 F 8 O at that time.Little is known of the c-C 4 F 8 O properties related to gas phase loss processes in the atmosphere.As part of the above-mentioned evaluations, a greenhouse warming potential (GWP ) is cited at a value on structural analogies to octafluorocyclobutane (c-C 4 F 8 ), for which the GWP is known (3M company-internal analysis cited byPruette et al. (2000)).A Material Safety Data Sheet for PFG-3480 (trade name for c-C 4 F 8 O), lists a GWP of 13 900 (100-yr time horizon) and a lifetime of <4 000 years(3M Company, 2007).It is unknown to us how these results were obtained.for c-C 4 F 8 O.
experiments to determine the infrared and ultraviolet (UV) absorption spectra of c-C 4 F 8 O, and the rate coefficient for the O( 1 D) + c-C 4 F 8 O reaction to estimate the atmospheric lifetime and GWP of c-C 4 F 8 O. 2 Methods 2.1 c-C 4 F 8 O in air samples 2.1.1Measurements of c-C 4 F 8 O in archived and ambient in situ air

✿✿✿✿✿✿✿✿✿✿✿ ( 33 •
N-53 • N) ✿ mostly using oil-free diving compressors.Two firn air samples were also analyzed, which were collected at the Aurora Basin North (ABN) site in Antarctica (71.1 • S, 111.4 • E).The site is located 550 km inland from Australia's Casey station, at 2710 masl and has a low mean annual air temperature of −44 • C. Samples were collected in December 2013; those for the halocarbon measurements were collected into internally electropolished stainless steel containers using a 2-stage teflon-coated viton diaphragm pump.
s functions from both the firn model and AGAGE 12-box atmospheric model described above to relate firn and tropospheric mole fraction to c-C 4 F 8 O surface emissions.The Green's functions from the 12-box model were calculated using a constant distribution of emissions into the four zonal boxes at the surface, and for this we used the relative contributions 0.675, 0.325, 0.0, and 0.0, in the northern-most to southern-most zonal bands.Results are fairly insensitive to emissions distributions that have all ✿✿✿✿ most emissions in the Northern Hemisphere (see Supplement).The characteristics of sparse atmospheric, firn and ice core data necessitate the use of constraints on the inversion to avoid unrealistic oscillations in the reconstructed mole fractions or negative values of mole fraction or emissions.The inversion uses non-negativity constraints and favors relatively small changes in annual emissions between adjacent years over large, unrealistic fluctuations.A prior emissions history is needed as a starting point for the inversion, ; ) ) is used to find the solution that minimizes a cost function consisting of the model-data mismatch

A
the firn model parameters through the use of an ensemble of firn Green's functions.2.3 Laboratory studiesLaboratory studies to measure the infrared and UV spectra of c-C 4 F 8 O and the rate coefficient for the O( 1 D) + c-C 4 F 8 O reaction were conducted at the Chemical Sciences Division Laboratories at the National Oceanic and Atmospheric Administration (NOAA), Boulder, Colorado, USA.The apparatus and methods used in this work are described separately below.is the measured absorbance at wavelength λ, I(λ) and I 0 (λ) λ) are the measured light intensities with and without the sample present in the absorption cell, respectively, L ✿ L ✿ is the optical absorption path length, σ(λ) ✿✿✿✿ σ(λ) is the infrared or UV cross section of c-C 4 F 8 O, and [c-C 4 F 8 O] is the concentration of c-C 4 F 8 O.In total, 11 independent absorption spectrum measurements were used in the linear least-squares fit.The c-C 4 F 8 O concentration was determined using the ideal gas law and absolute pressure measurements of either the pure compound or of a dilute mixture of the compound in a helium (He) bath gas.The c-C 4 F 8 O sample was obtained from SynQuest Laboratories (Inc., Florida, USA, 99 % purity).For the experiments described below, c-C 4 F 8 O was introduced into the absorption cells as a pure sample or in a dilute mixture prepared off-line.The dilute mixtures of c-C 4 F 8 O in a He (UHP, 99.999 %) bath gas was ✿✿✿✿ were prepared manometrically in a 12 L Pyrex bulb with an estimated accuracy of ∼1 % were measured with 100 Torr and 1 000 Torr (130 and 1 300 hPa, respectively) capacitance manometers.Quoted uncertainties are 2σ.
of c-C 4 F 8 O in He with a 0.00180 mixing ratio.The c-C 4 F 8 O concentration used in the absorption measurements was in the range 1.75×10 15 to 2.34×10 16 molecule cm −3 .Integrated band strengths (IBS) were obtained from the measurement of 11 individual IR spectra.The UV absorption spectrum of c-C 4 F 8 O was measured at 296 K using a 0-coupled device (CCD) detector.The collimated output of a 30 W deuterium lamp passed through a 100 cm long and 2.5 cm diameter Pyrex absorption cell with quartz windows.Spectral measurements were made over the wavelength region 200-350 nm.The wavelength scale of the spectrometer was calibrated using the emission lines from a low-pressure Hg pen-ray lamp.c-C 4 F 8 O was added to the absorption cell in pure form from the original sample.Measurements were performed over a range of c-C 4 F 8 O concentrations from 2.51×10 18 to 2.16×10 19 molecule cm −3 .Eleven independent UV absorption spectrum measurements were used in the final linear least-squares fit.
, which was 100 cm long and with a 2.2 cm internal diameter, was coupled with a Teflon circulating pump to an absorption cell where the loss ✿✿✿✿✿ losses of c-C 4 F 8 O and CHF 3 was ✿✿✿✿ were ✿ measured using FTIR spectroscopy.The FTIR absorption cell was equipped a multi-pass cell (485 cm path length) with KBr windows.Spectra were recorded in 100 co-adds at a spectral resolution of 1 cm −1 .

✿
photolysis ✿✿✿✿✿✿✿ lifetime, ✿✿✿✿✿✿✿✿ although ✿✿✿ the ✿✿✿✿✿✿✿ lifetime ✿✿✿✿✿✿✿✿✿ dependence ✿✿✿ on ✿✿✿ the ✿✿✿✿ cross ✿✿✿✿✿✿✿ section ✿✿✿✿ value ✿✿ is ✿✿✿ not ✿✿✿✿✿ linear ✿✿✿✿ due ✿✿ to ✿✿✿ the ✿✿✿✿✿✿ lifetime ✿✿✿✿✿✿✿✿✿✿ dependence ✿✿✿ on Hemispheresamples show higher mole fractions compared to the CGAA at similar times, suggesting predominant Northern Hemisphere emissions.c-C 4 F 8 O air samples fit well into the CGAA record with the older sample at slightly lower mole fraction than the oldest CGAA impossible to further pin down the first appearance of this compound in the atmosphere and the exact course of the abundance until ∼1980 because our knowledge of c-C 4 F 8 O prior to the CGAA is based on only one firn ✿✿✿✿✿✿ sample measurement with air spanning several decades (see calculated Green's functions in the Supplement). of the two firn air samples in the canisters is much shorter than those of the older CGAA samples, stability of c-C 4 F 8 O in the CGAA tanks and confirms that the observed ✿✿✿✿✿✿✿✿✿✿ multidecadal ✿ record is not a simple artifact of degradation of c-C 4 F 8 O in canisters over time.In situ measurements at Aspendale, which are available on a regular measurement basis since February 2017, show constant abundances ✿ a ✿✿✿✿✿✿✿ constant ✿✿✿✿✿✿✿✿✿ abundance of c-C 4 F 8 O at ∼74 ppq.This lack of growth is an indication of currently very small, if at all any, emissions of this compound.Also, pollution events are absent from this urban in situ record within the precision of these measurements, suggesting that c-C 4 F 8 O is not emitted within the airmass footprints of the site.the global mean abundance of c-C 4 F 8 O was 11 ✿✿ 10 ✿ ppq, cumulative emissions had reached 0.4 ✿✿✿✿ 0.38 kt.For the time after ∼1980, when observations became more frequent, emissions were 0.02-0.03kt yr −1 for about a decade.From the mid 1990s, emissions increased strongly to a maximum of 0.16 ✿✿✿✿ 0.15 (±0.04, 2 σ) kt yr −1 in 2004.Surprisingly, emissions have declined sinceand have reached 0.01 , kt.If scaled with the GWP on a 100-yr time horizons ✿✿✿✿✿✿ horizon, laboratory experiments of atmospheric loss processes and first atmospheric observations of c-C 4 F 8 O, a persistent greenhouse gas not regulated by the Montreal and Kyoto Protocols.We measured infrared and UV absorption spectra of c-C 4 F 8 O, and the rate coefficient for the O( 1 D) + c-C 4 F 8 O reaction.These experimental results suggest that c-C 4 F 8 O is an atmospherically persistent trace gas with an atmospheric lifetime of >3 000 years.In addition, its strong absorption in the "atmospheric window" results in a very high radiative efficiency, and when combined with the long atmospheric lifetime, yields a high global warming potential ✿✿✿✿ GWP ✿ of 12 000 (100-year time horizon), which is exceeded by only a few other synthetic greenhouse gases.We show an increase of c-C 4 F 8 O in the atmosphere to present mole fractions of ∼74 ✿✿ 75 ✿ ppq.Emissions, which were derived from these observations, have strongly declined after a peak in 2004.The reason ✿✿✿✿✿✿ reasons for this recent decline, and whether this is only a temporary feature, remains result of the industry's choice for alternative substances for chemical vapor chamber cleaning, which is ✿✿✿ we assumed to have been its primary use in the last two decades.However, even if emissions were completely halted, it will, due to the very long lifetime of thousands of years for the compound c-C 4 F 8 O to be removed from the atmosphere.Data ✿✿✿✿✿✿✿✿✿✿ availability: ✿✿✿✿ Data used in this study are available from the Supplement.Mention of trade names or commercial products does not constitute an endorsement or recommendation by NOAA for use ✿✿✿ for
c-C 4 F8O is the molar weight of c-C 4 F 8 O , F 8 O is a potent radiative forcing agent due to the combination of its high radiative efficiency and long atmospheric lifetime.Atmospheric observations and emissions of c-C 4 F 8 OWe observe a general increase of c-C 4 F 8 O in the atmosphere over the sample period starting in 1978 (Fig.4).c-C 4 F 8 O was detectable in all samples but abundances were ✿✿✿for