Spectroscopic Detection of Stratospheric Hydrogen Cyanide

A number of features have been identified as absorption lines of hydrogen cyanide in infrared spectra of stratospheric absorption obtained from a high-altitude aircraft. Column amounts of stratospheric hydrogen cyanide have been derived from spectra recorded on eight flights. The average vertical column amount above 12 kilometers is 7.1 � 0.8 x 1014 molecules per square centimeter, corresponding to an average mixing ratio of 170 parts per trillion by volume.

entirely well mixed; some inclusions in carbonaceous chondrites have a record of addition of small amounts (-0.1 percent) of heavy elements with nonsolar isotopic composition [reviewed by T. Lee, Rev. Geophys. Space Phys. 17, 1591 (1979)]. These samples are quite rare, however, and for the most part the nebula appears to have been well mixed in heavy elements to better than 0.01 percent. The evidence for "07Pd in Santa Clara (10) suggests that this meteorite received any late addition of recently synthesized elements. The limit for SHE's in Santa Clara can therefore be related to average solar system material, but the possibility of rare materials with higher amounts cannot be excluded. 22. A. Kracher, J. Willis, J. Wasson, Geochim.
We have analyzed high-resolution solar spectra at wavelengths near 3 p.m taken with a spectrometer aboard a Sabreliner jet aircraft flying in the lower stratosphere. This spectral region contains the strong V3 band of HCN, with band center at 3311.48 cm-'. A number of weak lines at the position of the HCN lines were identified. This spectral region also contains lines of H20 and N20 as well as some unidentified lines. The HCN lines are weak, so that the signalto-noise ratio in a single spectrum does not permit identification of HCN, but by averaging a large number of individual scans and observing lines from P,6 to R16 we have made a positive identification. To our knowledge, HCN has not been observed previously in the stratosphere or unpolluted troposphere.
The method of obtaining and analyzing stratospheric spectra has been reported (1). A solar tracking system directs sunlight into a high-resolution (0.06 cm-' full width at half-maximum, apodized) Fourier transform spectrometer. The instrumentation is flown aboard the Sabreliner at times near sunset or sunrise to enhance the absorption path length. Figure 1 shows a portion of the measured spectrum in the region containing the P branch of the v3 band of HCN. The HCN lines and major lines due to water are indicated. This spectral region has not been extensively studied in the past. To establish the presence of HCN absorption in the stratospheric spectrum, a wider portion of the spectrum was examined for HCN V3 lines with the line positions of Rank et al. (2). Although Table 1. Lines in the V3 band of HCN used for analysis. The line positions are from Rank et al. (2) and the intensities are based on measurements by Jaffe (6); the band intensity inferred from the line intensities given by Jaffe agrees within 3 percent of that from Hyde and Hornig (7).   Table 1 with their line intensities measured by Jaffee (6). The intensities are estimated to be accurate to + 20 percent or better, and they agree well with band intensities given by other investigators (7,8). To determine the amount of absorber, synthetic spectra are calculated (9) to match the observed spectra. In the absence of better information, we assumed a uniform mixing ratio of HCN above the aircraft (10). Since the lines are all weak, any other distribution would produce the same absorption for the same line-ofsight amount of HCN; any errors introduced by this assumption in the conversion of line-of-sight amounts to column amounts are negligible compared to the noise. The temperature profile of the U.S. Standard Atmosphere was used. The upper curve in Fig. 1 is a calculated spectrum based on line parameters for H20, C02, and N20 from the AFGL line compilation (4) and the HCN line parameters in Table 1. Table 2 gives details of the flight spectra that were analyzed to determine amounts of HCN. Flights were made at latitudes from 50 to 50°N in winter and summer. The errors in the determination of HCN column amounts are largely due to noise in the spectra; systematic errors due to assumptions of distribution, temperature effects, errors in line intensities or broadening coefficients, and so on are generally much smaller. To improve the signal-to-noise ratio, we averaged as many spectra as possible, even when this covered a significant range of air mass, 0036-8075/81/1016-0333$01.00/0 Copyright © 1981 AAAS and used the mean air mass for conversion to a vertical column amount. The standard deviations of the columnn anounts measured in the six lines are shown in Table 2. The mean of all the cb.servations, weighted by the inverse square of the individual-standard deviations, gives a column amount of 7.1 x 1014 molecules per square centimeter in a vertical column above 12 km, with a standard deviation of the mean of 0.81 x 104 cm-2. This quantity corresponds to an average mLxing ratio of 170 ± 20 parts per trillion by volume above 12 km. Although the measurements cover the latitude range 5°to 50°N in winter and summer, the precision of the individual points is not high enough to allow us to observe any trend with latitude. Sinilarly, although there is sonie indication that the amount is lower in the sunrise spectrum than at sunset, there are not enough sunrise data to make a definitive statement about diur-nal variability; no diurnal variation is expected, as discussed below.
The origin of HCN and its role in atmiospheric chemistry are not completely clear. Although it was not detected previously in the stratosphere or free troposphere, HCN is a known combustion product (11) and can be produced by microbes and plants. Atmospheric HCN apparently originates at ground level from these sources (12). Although mechanisms exist for its in situ atmospheric production (12), they appear to be inadequate to explain the concentrations we observed. Similarly, direct injections of HCN from high-flying aircraft (13) are too small to account for our data. Once in the atmosphere, HCN is not very reactive because of its strong bonds. Because the bond dissociation energy D(H-CN) = 119.9 kcal/mole there are few, if any, possible atmospheric hydrogen-abstraction reactions; for instance, reaction with OH is slow (14) and endo- thermic. Although the reaction HCN + CIO --HC1 + NCO is exothermic, one expects four-centered reactions to be extremely slow in the atmosphere because of -their high activation energies (15). Also, HC(N is photodissociated only by short-wavelength (< 200 nm) light (16) and its Henry's law coefficient implies that clouds and rain are ineffective scavengers of HCN vapor. Accordingly, the atmospheric residence time of HCN is measured in years, and so its vertical distribution is largely controlled by transport and no diurnal variations should be observable in the stratosphere. These points are discussed more quantitatively in (12). The concentration in the troposphere should be as high as that in the stratosphere, but spectroscopic detection there will be difficult due to interference by water vapor.
Recently, the presence of CH3CN in the stratosphere was inferred from positive-ion mass spectra (17), and CN-(hydrated) appears faintly in stratospheric negative-ion spectra (18). The decomposition of stratospheric CH3CN cannot produce much HCN because CN reacts quickly with 02 to form NCO, which absorbs visible light (12), cleaving the C-N bond. Also, while CN-and HNO3 can react to form HCN, this HCN source appears small. Finally, because the atmosphere cannot synthesize CH3CN from HCN, it seems likely that the atmospheric chemistries of CH3CN and HCN are uncoupled. Substance P in Principal Sympathetic Neurons: Regulation by Impulse Activity Abstract. Regulation of the putative peptide neurotransmitter substance P was examined in the superior cervical sympathetic ganglion ofthe neonatal rat. Surgical decentralization (denervation) of the superior cervical ganglion increased ganglion substance P content. In cultured ganglia, the amount ofsubstance P increased more than 50-fold after 48 hours, and this rise was dependent on protein and RNA synthesis. Veratridine prevented the increase in substance P in vitro, and tetrodotoxin blocked the veratridine effect; this suggests that sodium influx and membrane depolarization prevent substance P elevation. Immunohistochemical analysis of cultured ganglia indicated that substance P was present in the perikarya ofprincipal sympathetic neurons and in ganglionic nerve processes. Transsynaptic impulses, through the mediation ofpostsynaptic sodium influx, may decrease substance P in sympathetic neurons.
Traditional concepts of neuronal specificity and brain function have been dramatically altered by recent work suggesting that peptides act as neurotrans-mitters (1,2). It is now apparent that peptidergic neurons are distributed throughout the neuraxis and that peptides and well-recognized transmitters, such as norepinephrine, may coexist in the same neurons (2). In order to examine the functional implications of these observations, we studied peptidergic expression and metabolism in a relatively simple neuronal structure, the rat superior cervical sympathetic ganglion (SCG).
Traditional teaching maintains that sympathetic ganglion neurons use only norepinephrine or acetylcholine as transmitters and that nerves innervating ganglia are cholinergic (3). However, the recent demonstration of putative peptide transmitters in sympathetic ganglia (4,5) suggests that the biochemical organization of the sympathetic nervous system is more complicated. For example, the undecapeptide substance P has been detected in ganglion nerve fibers (4), and application of substance P to sympathetic neurons elicits membrane depolarization and neuronal discharge (6). Moreover, since the peptide is released from ganglia by a high potassium stimulus in a calcium-dependent manner, substance P appears to subserve a physiologic role in sympathetic ganglia (7).
Recently, we found that surgical decentralization (denervation) of the SCG in the adult rat, or pharmacological blockade of transmission, increased sub-