Technical Comments

Comment on “Missing gas-phase source of HONO inferred from Zeppelin measurements in the troposphere”

See allHide authors and affiliations

Science  19 Jun 2015:
Vol. 348, Issue 6241, pp. 1326
DOI: 10.1126/science.aaa1992


Li et al. (Reports, 18 April 2014, p. 292) proposed a unity nitrous acid (HONO) yield for reaction between nitrogen dioxide and the hydroperoxyl-water complex and suggested a substantial overestimation in HONO photolysis contribution to hydroxyl radical budget. Based on airborne observations of all parameters in this chemical system, we have determined an upper-limit HONO yield of 0.03 for the reaction.

The hydroxyl radical (OH) is the most important oxidant in the atmosphere, affecting the steady-state concentrations and lifetimes of most gaseous pollutants (1). Ground-based field studies have demonstrated that nitrous acid (HONO) is an important or even a major OH precursor via its photolysis (2, 3). However, Li et al. (4) proposed a new HONO source through the reaction between nitrogen dioxide (NO2) and the hydroperoxyl-water complex (HO2·H2O) (R1A). Because this source mechanism consumes an HO2 radical and NO2 (defined as an internal mechanism), they suggested that HONO photolysis might not be an important net hydrogen oxide radical (HOx) source, as previously believed.

Embedded ImageEmbedded Image

Li et al. (4) presented some high-quality HONO observations in the residual layer (RL) during the morning hours, one of only a few airborne HONO measurements (46). Decoupled from the ground surface emission and upper troposphere convection, the RL is an ideal environment to study the nighttime-to-daytime evolution of HONO chemistry within an air mass. High HONO concentration observed after sunrise, ~150 parts per trillion (ppt), indicates a high HONO production rate in the air mass. Their model simulation suggested that an external HONO source, defined as HONO production from a mechanism without consuming HOx radicals and oxides of nitrogen (NOx = NO + NO2), would lead to an overprediction of NOx concentration after sunrise, whereas the internal source based on reaction R1A well reproduced both HONO and NOx concentrations in the RL. The authors thus recommended the internal HONO source R1A over the external ones for sustaining the high HONO concentrations, with a HONO yield (α) of 1. However, there have been no reports of direct laboratory evidence suggesting HONO formation via reaction R1A, and the assumed α value of 1 seems unreasonably high.

We conducted HONO measurements on board National Science Foundation’s National Center for Atmospheric Research (NSF/NCAR) C-130 research aircraft during the 2013 summer NOMADSS (Nitrogen, Oxidants, Mercury and Aerosol Distributions, Sources, and Sinks) field study. HONO was measured by two long-path absorption photometric (LPAP) systems (6, 7); NO and NO2 were measured by a four-channel chemiluminescence instrument (8); HO2 was measured by a chemical ionization mass spectrometer (9); and photolysis frequency of HONO, JHONO, was derived from light measurement by a scanning actinic flux spectroradiometer (10). Figure 1, A to C, shows measurement data, including HONO, NO, NO2, HO2, water vapor (H2O), and HONO photolysis frequency (JHONO), during a 1-hour segment of research flight on 29 June 2013 at 800 m above ground over central Georgia, USA. The mean HONO (±SD) concentration was 18 (±6) ppt. Elevated HONO concentrations were observed in the power plant plumes, at 19:01, 19:09, and 19:24 UTC (A, B, and C in Fig. 1A). The maximum HONO concentration in these plumes was 37 ppt, much lower than the ~150 ppt reported by Li et al. (4), even though the NO2 and HO2 levels were higher. Figure 1D compares the HONO photolytic loss rate and the upper limit for a hypothetical HONO formation rate from reaction R1A. The calculated HONO formation rate is at least 10 times greater than the observed HONO photolytic loss rate. This suggests that HONO contribution from reaction R1A as assumed by Li et al. (4) has been grossly overestimated since the photolytic loss is the dominant sink for HONO.

Fig. 1 Time series.

Time series of (A) HONO and JHONO, (B) HO2 and H2O, (C) NO2 and NO, and (D) HONO photolysis loss rate and the upper-limit HONO production rate from reaction R1A. The upper-limit HONO production rate from reaction R1A is calculated under the same assumption as that of Li et al. (4). Three power plant plumes were encountered, as indicated by A, B, and C in (A). The flight altitude was 800 m above ground level.

Indeed, an upper-limit α of 0.03 is calculated for reaction R1A through Eq. 1, assuming it is the sole source to counterbalance the HONO photolytic loss in the plume C (Fig. 1A)Embedded Imagewhere [HONO], [HO2·H2O], and [NO2] are concentrations of HONO, HO2·H2O, and NO2, respectively. JHONO is the HONO photolysis frequency calculated from irradiance measurement; k is the overall rate constant of reaction between HO2·H2O and NO2 (i.e., R1A + R1B). Other HONO sources (11)—including OH radical and NO reaction, NO2 heterogeneous reactions on aerosol particles, and the photolysis of absorbed nitrate—are ignored here. Therefore, even the low value of 0.03 may still greatly overestimate the true α value.

With such a small α, this internal mechanism (R1A) is not able to sustain the HONO level observed in Li et al. (4), and other internal and external sources need to be considered. The HONO source calculated in Li et al. (4) was 4.05 × 105 × JHONO (ppt·s·h−1), indicating a photochemical process with a precursor of substantially long lifetime, such as the photolysis of particulate nitrate (6, 12). Although particulate nitrate loading was not measured in Li et al. (4), the aerosol surface area density, up to 1.8 × 10−4 m2 m−3, suggests that a relatively high aerosol loading, and perhaps high particulate nitrate loading in the industrial region of the Po Valley, Northern Italy, could be an explanation for their observation.

Li et al. (4) reported that NOx concentration would be overpredicted in their simulation if an external HONO source was considered. We argue that their model may have underpredicted the NOx loss rate. Model parameterization should be updated and be constrained by measurements to accurately assess important NOx sink processes, such as formation of organic nitrates from alkylperoxyl radicals and NO reaction (13), peroxyacyl nitrates (PANs) from peroxyacyl radicals and NO2 reactions (14), and bromine nitrate from bromine oxide and NO2 reactions (15).

Finally, HONO will remain an important net OH precursor, as demonstrated by many field studies (2, 3), because HONO formation from reaction R1A is negligible, with an α value lower than 0.03.

References and Notes

  1. Acknowledgments: This research is funded by NSF grants AGS-1216166, AGS-1215712, and AGS-1216743. We appreciate the help and assistance from fellow scientists and NCAR’s C130 crews during the NOMADSS field study. The data are available in our project data archive ( Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of NSF.
View Abstract

Navigate This Article