Projects

A proposal for an intercomparison exercise of ambient measurements of peroxyacyl nitrates (PANs) and other organic nitrates

Frank Flocke and Elliot Atlas, ACD
Submitted through: Daniel S. McKenna, ACD Division Director

Abstract

A comparison of trace gas measurement techniques between laboratories provides the atmospheric sciences community with an opportunity to evaluate trace gas data from diverse measurement campaigns. Such evaluation is necessary to analyze available measurements in an integrated and globally consistent data set. Comparisons have been conducted for a variety of trace gases, including non-methane hydrocarbons, formaldehyde, sulfur species, and some reactive nitrogen compounds. Here we propose to conduct an international intercomparison campaign of techniques for the ambient measurement of PANs and related organic nitrates. No significant international comparison exercise for PANs has been conducted to date. Such a comparison is required, especially on an international scale, because PANs are a class of compound that is significant in the chemistry of the global troposphere. Measurement of PANs and other organic nitrates provides information on the transport, transformation, and recycling of reactive nitrogen that is related to oxidant formation and loss throughout the troposphere. Other organic nitrates, such as alkyl nitrates, also play a significant role in nitrogen transformation and recycling, especially in regions of high VOC and/or NOx emission. The campaign is targeted for late summer 2003. Startup funds for the preparation and facilitation of the intercomparison campaign as well as for the organization of a workshop for evaluation of results are requested.

Introduction and Rationale

There is an increasing emphasis on regional and global scale modeling using data assimilation to study atmospheric chemical processing and transport to assess anthropogenic impacts on the troposphere, and to predict future changes in tropospheric composition, oxidizing capacity, climate impacts, etc. Consequently, there is a need for accurate and consistent data sets to be used for modeling on both the process and the regional and global scales. Especially in cases where different and new analytical methods are being used in the atmospheric chemistry community to measure a specific molecule or group of compounds, intercomparison exercises are crucial to ensure data consistency and accuracy. Often, intercomparisons uncover unsuspected interferences or other problems in different measurement techniques. In addition, known interferences of certain methods can be examined and quantified in a controlled environment (variables like humidity and pressure can be varied) and the method can be assessed for its suitability, for example, for aircraft measurements. Intercomparison campaigns ultimately improve data quality and the usefulness of data sets for atmospheric modeling and prediction.

Several, very successful, intercomparison exercises have been coordinated at NCAR, such as NOMHICE (an ongoing international non-methane hydrocarbon intercomparison), an intercomparison for formaldehyde measurements in 1995, and IPMMI (an intercomparison for atmospheric radiation measurements), and elsewhere. The CITE and CITE II aircraft intercomparison measurements of reactive nitrogen compounds and others, and PRICE (a comparison of peroxy radical measurement techniques) are to be listed here. Some informal, small scale comparisons (i.e. during SOS, TexAQS 2000 and, recently, TRACE-P) have been done; however, a comprehensive, international intercomparison campaign for PANs and other organic nitrates has never been conducted. Therefore, and in light of recently developed new methods for the measurement of PANs and organic nitrates, it seems necessary and timely to propose an intercomparison exercise for this important group of photochemical oxidants.

PANs in the atmosphere

Peroxyacetyl nitrate {PAN; CH₃C(O)OONO₂} is an important member of the reactive nitrogen family, and it is present throughout the global troposphere. Among a number of peroxyacyl nitrates {PANs; RC(O)OONO₂} presently known to exist, PAN is the most abundant in ambient air.  Mixing ratios range from a few pptv or less in the remote marine boundary layer to several ppb in heavily polluted urban regions. Other PANs have been observed, but typically at concentrations much less than PAN. These less abundant PAN-type compounds are peroxypropionyl nitrate {PPN, C₂H₅C(O)OONO₂}, peroxyisobutyryl nitrate {PiBN, (CH₃)₂CHC(O)OONO₂}, peroxymethacryloyl nitrate {MPAN, CH₂=C(CH₃)C(O)OONO₂}, peroxyacryloyl nitrate {APAN, CH₂=CHC(O)OONO₂}, and peroxybenzoyl nitrate {PBzN, C₆H₅C(O)OONO₂}. Other complex PANs are predicted by photochemical oxidation models, but they have yet to be measured in the ambient atmosphere. PANs are formed during the photochemical oxidation of non-methane hydrocarbons in the presence of NO_x­, which makes them by-products of the rate-limiting steps in photochemical ozone formation in the troposphere. This makes PAN species important for several reasons. First, the variety of the hydrocarbons involved in the photochemistry is reflected in the relative abundance of the different PAN species. For example, PAN arises from almost all non-methane hydrocarbon species, whereas PPN originates mainly from longer chain anthropogenic hydrocarbons like alkanes and alkenes and some biogenic hexene derivatives. MPAN, however, is formed primarily during the photooxidation of isoprene, which has predominantly biogenic sources such as trees and phytoplankton. Hence the absolute and relative concentrations of different PANs provide an excellent tracer for this chemistry and the relative contributions of anthropogenic and biogenic hydrocarbon emission.

A second important aspect to PANs is their role in the budget and transport of reactive nitrogen species in the atmosphere. Throughout most of the troposphere ozone production rates are controlled by the availability of nitrogen oxides (NO_x). NO_x levels are determined by emissions, deposition, and, most importantly, cycling between different reactive nitrogen reservoir species. PAN type compounds often constitute a major fraction of available reactive odd nitrogen, NO_y, which is defined as the sum of NO_x and all NO_x-product species (e.g. nitric acid, nitrous acid, PANs, alkyl nitrates, etc.). PANs are in equilibrium with their precursors, the respective peroxyacyl radicals and NO₂. Net PAN losses primarily occur via the reaction of the peroxyacyl radical with NO or with HO₂ or other organic peroxy radicals and, to a lesser extent, via photolysis and reaction with OH (with the OH reaction becoming more important in the case of MPAN). Loss of PANs via thermal dissociation has a very strong temperature dependence. Lifetimes of PAN vary from less than an hour at temperatures of 300K or higher to months at temperatures characteristic of the upper troposphere. Thus, PANs are quite stable in the mid- and upper troposphere, and they can be transported over long distances from polluted continental regions into the remote troposphere such as over the Pacific Ocean. PANs therefore act as a temporary reservoir for NO_x and, through long range transport and subsequent release of NO_x especially in the lower troposphere, PAN may control the photochemical production of O₃ in much of the remote marine troposphere. Since the mid-80s, a number of studies of PAN in remote marine environments such as over the Pacific or Atlantic Oceans have shown layers of polluted air at mid- to upper altitudes over the oceans, in which PAN is an abundant, and sometimes the most abundant, odd-nitrogen species. The marine boundary layer appears to be depleted in PAN relative to typical continental sites, indicating loss mechanisms likely resulting in reformation of NO_x. This may represent one of the most important mechanisms by which NO_x is transported globally and may control the NO_x budget and therefore the photochemical ozone production throughout large portions of the remote lower troposphere.

As a consequence, accurate and precise measurements of PAN and related compounds are crucial to the understanding of the NO_x budget and ozone photochemistry of the troposphere. Accurate and consistent measurements of PANs are necessary to characterize polluted air transported into remote areas, the photochemical evolution of plumes from pollution sources, and the influence of NO_x release from PANs decomposition on the photochemistry of the remote troposphere, for example over the Pacific Ocean. In continental regions, PANs and complex organic nitrates resulting from oxidation of biogenic hydrocarbons may be important in the atmosphere-biosphere exchange of nitrogen. Thus, measurements of PANs provide information on photochemical processes occurring closer to source regions over the continent, as well as their impact on the global troposphere.

Current State of the Art of PAN Measurements

Measurements of PAN have most often been accomplished by gas chromatography with electron capture detection. This method is currently the best developed and most proven way to measure PAN and its analog species, PPN, PiBN, and MPAN. The method has been improved by the application of capillary chromatography and some instruments can now provide an analysis of the above four species in 5 minutes or less. Gas chromatography/negative-ion, chemical-ionization mass spectrometry (GC/NICI MS) is a new technique that has recently been reported as a selective method for PANs. Gas chromatography with a luminol-based detector has also been used for rapid and selective detection of PAN.  A new and promising method for the semi-continuous ambient measurement of PANs is the proton-transfer mass spectrometry (PTR-MS) technique.  Laser-induced fluorescence of NO₂ from thermal degradation of PANs or alkyl nitrates is another technique that attempts to measure total amounts of different classes of reactive nitrogen species (including organic nitrates). A promising CIMS technique is currently under development at Georgia Tech. Please see table 1 for details on techniques and interested participants. With the advent of a variety of new techniques and extensive application of more “proven” techniques, it is an excellent opportunity to bring together the various instruments for a critical comparison exercise.

As in any trace gas analysis, it is not only the actual analytical technique that contributes to the uncertainty of the measurement, but also the method of standardization. One critical part of the analysis process is the production of a gas phase standard that is of known concentration and high stability and purity. Such a standard is required if accurate measurements are to be obtained that can be compared to other NO_y species measurements, and to assure comparability of measurements acquired at different places and different times. Thus, a significant component to a comparison of PAN measurement techniques involves an evaluation of standard reference materials for PANs and related compounds.

PANs are sufficiently unstable that conventional gas standardization techniques, similar to those for hydrocarbons or halocarbons, are not applicable, and relatively sophisticated techniques have been developed and need to be compared. The synthesis of PAN-type compounds from the peroxy carboxylic acid by strong acid nitration has been shown to yield PANs of high purity in the cases of PAN, PPN and PnBN (Nielsen et al., 1982; Gaffney et al., 1984; Roberts, 1990). These compounds can be extracted into relatively high-molecular weight alkane solvents and placed in diffusion devices. As long as these solutions are kept at 0^oC or below, they can last for up to 1 week without significant deterioration. Some drawbacks of this method are: a) independent calibration is required, either spectroscopic or by total NO_y measurement, b) alkane solvents that have been used are liquids at 0^oC, leading to mechanical instability in aircraft applications, and c) compounds such as PiBN and MPAN cannot be synthesized with sufficient purity to be used directly in this method.

An on-line photochemical method has been developed in ACD for the calibration of PAN. The method consists of the photochemical production of peroxyacetyl (PA) radicals from the 285nm photolysis of acetone in the presence of O₂ and CO. A small, accurately measured flow of an NO calibration standard is added to the gas stream and efficiently (>90%) converted first to NO₂ then to PAN. The main advantages of this technique are that the PAN calibration is referenced to the initial NO standard concentration, a stable quantity that can be compared to the calibrations of other NO_y measurements, and there is the potential to extend the technique to other PANs.

Proposed work

Ambient measurements would be compared by concurrent sampling from a continuously pumped and flow controlled community sampling manifold, which would be located centrally to minimize sampling line lengths from the manifold to the individual trailers housing the instruments. In addition to intercomparing ambient measurements, PAN and nitrate standards and various test mixtures (for example species which are suspected to potentially interfere with some detection methods) would be introduced into the manifold. The manifold additions would be done while simulating a wide range of conditions such as pressure and relative humidity levels. Because of the thermal instability of PANs, ambient air can be easily scrubbed of PAN-type compounds by using a simple heat exchanger, and such scrubbed ambient air as well as zero air from a zero air generator will be used as the diluent matrix for the PAN standards. It is planned that this will be a blind intercomparison and the results of the participating groups will be submitted to and evaluated by an independent referee.

ACD, in close collaboration with the NOAA Aeronomy Lab, has excellent expertise in the preparation of standards for and in the operation and calibration of PAN instrumentation. We have three different, well characterized sources for PANs, diffusion tubes for the major PAN compounds (based on diffusion of pure PANs samples diluted in a tri/pentadecane mixture through a glass capillary), a photolytic source for each PAN and PPN (based on the photolysis of ketones in the presence of NO_x), and a preparative scale gas chromatograph, used to produce pure known amounts of the more unstable, unsaturated PAN homologues APAN and MPAN. All these methods have been used and intercompared within and between ACD and at the Aeronomy Lab and can be calibrated independently using a NO_x/y instrument.

Given the co-location of the various PAN instruments, and the fact that some of the instruments measure additional organic nitrate species, we propose as a secondary objective to compare measurements of other organic nitrates, including simple alkyl nitrates, and more complex multifunctional organic nitrates. This secondary objective is timely since there are more groups involved in the analysis of alkyl nitrates, and recent measurements have suggested a large fraction of unidentified organic nitrates in the ambient atmosphere. For example, measurements conducted at Blodgett Forest, California, carried out with the NO₂ LIF/thermal dissociation instrument suggest a considerable fraction of unidentified alkyl nitrate compounds exists at this site under photochemically active conditions and transport of polluted air to this heavily forested area. In addition, we are currently conducting smog chamber research on the formation of multifunctional alkyl nitrates from isoprene and other reactive hydrocarbons in collaboration with the Atmospheric Kinetics group in ACD. The proposed experiment places those with appropriate expertise and instrumentation in a time and place to begin to examine this issue. Table 1 lists interested participants and their measurement capabilities.

The targeted time frame for the campaign is summer 2003. The campaign would be conducted at a location that provides both the necessary infrastructure and a range of ambient conditions ranging from remote to polluted. The Mesa Lab and the NOAA site on the Enchanted Mesa above Boulder are very suitable sites, since they provide the necessary infrastructure and a wide range of air mass pollution levels. Nearby vegetation will provide local emissions of biogenic hydrocarbons, which are suspected to form the multifunctional alkyl nitrates and complex PANs mentioned above.

NOAA, NASA and NSF have been informed of the intent to conduct a PAN intercomparison exercise at NCAR and have responded very favorably. A letter of support from the NASA Program Director of the GTE program is attached. The NOAA Aeronomy Lab has expressed great interest and may be able to provide logistical support at the Enchanted Mesa site and provide supporting measurements such as NO_y, NO_x, and HNO₃. NSF responded favorably to the proposed undertaking but strongly encouraged NCAR co-sponsorship to the cost of the campaign. Therefore, the availability of internal funding for this project will greatly improve the chance of NSF committing to fund proposals from the university groups to participate in this exercise.

Relation of this proposal to the Director's Opportunity Fund Guidelines

1. NCAR Strategic Plan

This proposal is directly in line with many aspects of the NCAR strategic plan. NCAR would facilitate a community effort that will involve the university community as well as other federally funded research labs. The campaign will provide critical information about the comparability of existing data sets, which are used by global and regional models. The exercise will also allow the evaluation of new but unproven methods for the measurement of PANs and organic nitrates which will greatly enhance the measurement capability of the atmospheric chemistry community.

2. Research and Education

As outlined above, new measurement methods will be evaluated and the use of data sets in global and regional models will be enhanced. Education and training will be facilitated through the involvement of graduate students working in the university groups.

3. Underrepresented Groups

No specific effort is made to involve underrepresented groups in the proposed effort. However, such groups may be represented in some of the participating university collaborators.

4. Infrastructure enhancement

The contribution to new method evaluation as outlined under Point 1 is a significant enhancement research infrastructure with regard to instrumentation.

5. Enhancement of scientific and technological understanding

The results of this exercise will be made widely available through publication of the results on the NCAR web site as well as in peer-reviewed journals. The benefits for the atmospheric science community are outlined above.

6. Benefits to the society

Accurate and precise measurements of PANs and organic nitrates contribute to the enhancement of knowledge about local and regional ozone formation as well as the oxidizing capacity of the remote atmosphere. Both aspects are relevant with regard to air quality and global change issues.

7. Collaborations and university interactions

As outlined above, this exercise will involve a large number of university research groups as well as other federally funded institutions and the research instrumentation industry. It will provide an opportunity for these groups to compare their proven instrumentation and calibration practices as well as to evaluate newly developed methods and instruments under a wide range of conditions.

Time and Cost Plan

A steering committee (consisting of three to four leading scientists in the field of reactive nitrogen measurements) has been selected and we expect to hold a steering committee meeting in the fall of 2002 to plan the campaign. A science overview document and a campaign implementation plan will be submitted by the steering committee to NSF by fall 2002. US participants will propose directly to NSF to obtain support to cover their cost of participation. The campaign will take place in late summer 2003. A data workshop will be held in late 2003 or early 2004 to discuss results and to facilitate the publication of the findings. Funding is requested for a meeting of the steering committee to prepare and plan the campaign, to provide the infrastructure for the campaign (providing the necessary gases, acquisition of a zero-air generator, construction of an inlet manifold, preparing trailers, provide extra tubing and fittings) and to hold a workshop to discuss, evaluate and plan the publication of results. ACD will provide funding for supporting measurements such as NO, NO₂, NO_y, O₃, HNO₃, alkyl nitrates and other organic nitrates (Ridley, Atlas and Eisele groups).

List of Acronyms

TRACE-P Transport and Chemical Evolution over the Pacific Experiment (NASA, 2001)

NOMHICE Nonmethane Hydrocarbon Intercomparison Experiment (NCAR, 1991-present)

CITE Chemical Instrumentation Test and Evaluation Experiment (NASA, 1983-1989)

PRICE Peroxy Radical Intercomparison Experiment (Germany, 1993)

IPMMI International Photolysis Frequency Measurement and Model Intercomparison (NCAR,1998)

Table 1: Interested participants and their measurement capabilities and techniques

Group	Contact	Measurement Technique
Western Michigan University	Stephen Bertman	PAN, alkyl nitrates (GC[1])
University of California Berkeley	Ronald Cohen	NO2,organic nitrates, NOy (LIF [2]/thermal decomposition)
New YorkState Univ., Albany	KennethDemerjian	PAN (GC)
Georgia Tech	Greg Huey	HNO3 (PTR-MS [3]), PAN (CIMS[4]
University of Washington	Daniel Jaffe	PAN, PPN (GC)
Harvard University	William Munger	PAN (GC)
University of North Carolina	Harvey Jeffries	PAN, alkyl nitrates (GC)
Purdue University	Paul Shepson	PANs, org. nitrates (GC/Luminol[5])
Argonne National Laboratories	Jeff Gaffney	PAN (GC/Luminol)
Battelle	Chet Spicer	PANs
NOAA / ERL	James Elkins, Fred Moore	PAN (GC)
NOAA / AL	JamesRoberts	PANs, alkyl nitrates (GC)
NOAA/ AL	Tom Ryerson	NOx/y(CLD)
NASA / ARC	HanwantSingh	PANs (GC)
NCAR	Brian Ridley	NOx,y (CLD [6])
NCAR	Frank Flocke, Andy Weinheimer, Elliot Atlas	PANs (GC, fast GC) organic nitrates (GC/MS[7])
University of Ulm, Germany	Karl-Heinz Ballschmitter	PANs (GC)
RISOE, Denmark	Anders Feilberg	PAN (GC)
University of Patras, Greece	Sotirios Glavas	PAN (GC)
University of Innsbruck, Austria	Armin Hansel	PAN (PTR-MS)
DLR, Germany	Joerg Heland	PAN (GC,PTR-MS)
Alfred-Wegener-Inst., Germany	Hans-Werner Jacobi	PAN
University of Tokyo, Japan	Yutaka Kondo	PANs, NOxy
Forschungszentrum Jülich, GER	Ralf Koppmann	PAN (GC)
University of Thessaloniki, GR	Kostas Kourtidis	PAN (GC)
NILU, Norway	Terje Krognes	PAN (GC)
University of East Anglia, UK	William Sturges	PAN, alkyl nitrates
University of Paris 12, France	Pascal Parros	PAN (GC)
University of Munich, Germany	Bernhard Rappenglück	PAN (GC)
METCON, Inc.	Rainer Schmitt	PAN, PPN (GC)
University of Mainz, Germany	Jonathan Williams	PANs (GC,PTR-MS)