HAL Id: jpa-00227431
https://hal.archives-ouvertes.fr/jpa-00227431
Submitted on 1 Jan 1988
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
SPECTRAL OBSERVATIONS OF THE EXTREME ULTRAVIOLET ASTRONOMICAL BACKGROUND
RADIATION
S. Labov, S. Bowyer
To cite this version:
S. Labov, S. Bowyer. SPECTRAL OBSERVATIONS OF THE EXTREME ULTRAVIOLET AS-
TRONOMICAL BACKGROUND RADIATION. Journal de Physique Colloques, 1988, 49 (C1),
pp.C1-63-C1-66. �10.1051/jphyscol:1988111�. �jpa-00227431�
JOURNAL DE PHYSIQUE
Colloque C1, Suppl6ment au n03, Tome 49, Mars 1988
SPECTRAL OBSERVATIONS OF THE EXTREME ULTRAVIOLET ASTRONOMICAL BACKGROUND RADIATION
S. LABOV and S. BOWYER
Space Sciences Laboratory, University of California, Berkeley, CA 94720, U.S.A.
RESUME
Les observations faites prk6demment sur le fond diffus uv et dans les rayons x B basse 6nergies impliquent plusieurs composant de gaz chaud (@ a lo6 K) dans le medium interstellaire. Si ce gaz est pr6sent en large volume dans le gaz local, des raies d'6missions devraient €.we produites dans I'extrkme ultraviolet (EUV). k ce point, le fond E W a 6t6 d6tect6 par des instruments photom6triques mais il n'y a pas d'observations spectroscopique en dessous de 5206;. Nous avons constmit un spectrometre B incidence rasante pour 6tudier le fond EUV entre 80 et 6506; avec une r6solution de 10 B 301. Cet instrument fut lance par une fus6e sonde en 1986, et les donnCes demontrent la pr6sence de plusieurs raies d'6missions. Une analyse p r C l i i a i r e montre que lusieurs raies qui seraient attribue au gaz interstellaires ont 6t6
K
dbtectks, en plus des raies interplanCtaire de He I 584 et la raie geocornale He II 3046;. Dans cet article nous discu- terons l'analyse de ces raies et les contraintes que I'on peut donc imposer s u les conditions dans le gaz interstellaire local.
ABSTRACT
Observations in the far ultraviolet and soft x-ray bands suggest that the interstellar medium contains several components of high temperature gas ( l d to lo6 K). If large volumes of local interstellar space are filled with this hot plasma, emis- sion lines will be produced in the extreme ultraviolet (EUV). Diffuse EUV radiation, however, has only been detected with photometric i n s m e n u ; no spectral measurements exist below 5206;. We have designed a unique grazing incidence spectrometer to study the diffuse emission between 80 and 650b; with 10 to 30b; resolution. This inshument was successfully flown on a sounding rocket in April of 1986 and a preliminary analysis reveals several features. In addition to the expected interplanetary He I 58461 emission and the geocosonal He 11 3046; emission, other features appear which may originate in the hot ionized interstellar gas. These features are discussed along with the possible implications to the hot phase of the interstellar medium.
Introduction
Astronomical diffuse radiation in the extreme ultravioIet (EUV: hh-100 to 1000A) was first detected in the 114 to 1501 wavelength band with a wide field broad band photometer on a sounding rocket 111. Since then, additional broad band observations have confirmed thisdetection 121 and have set upper limits for the radia- tion at longer wavelengths in the EUV band 13, 4, 5, 6'.
It is generally believed that this background originates from a hot, tenuous gas in the interstellar medium The origin of this gas, however, is controversial, and a definition of its characteristics is crucial to advancing our understanding of the interstellar medium. In particular, detection of thermal emission lines from this radiation would confirm the existence of this gas and would pro- vide information on its nature.
We have developed and built a spectrometer to measure the diffuse extreme ultraviolet astronomical background /7/. Our spectrometer design features a fast optical system utilizing grazing incidence optics and a large solid angle. Grazing incidence is essential to maintain high efficiency at wavelengths below 4006;,
and the large product of solid angle and effective area gives the spectrometer high sensitivity to diffuse emis- sion.
The layout of the spectrometer is shown in Figure 1. Photons enter the instrument though a wire grid coi- limator which restricts the field of view in one dimen- sion to 40: while allowing 15' of sky to enter the instru- ment in the orthogonal direction. This wedge of light is then diffracted by an array of flat, blazed, reflection grat- ings at grazing incidence. Once diffracted, the light is focused in the spectral direction by an array of mirrors through thin film bandpass filters and onto a detector.
The measured resolution (AIM.) of the instrument as a function of wavelength is shown in Figure 2. The diffuse EUV spectrometer is actually b eseparate spec- trometers focusing on two different detectors. The medium (230 to 430di) and long (430 to 6506;) wavelength systems are similar, and focus on the same detector. The short (80 to 230A) wavelength system uses twice as much area as either the medium or long wavelength spectrometen and has its own detector. The instrument is described in more detail in Labov et al. 171.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1988111
C1-64 JOURNAL
DE
PHYSIQUEWIRE GRID COLLIMATOR
DJXSRACTION GRATINGS
PLATINUM COATINGS
GLASS MIRRORS
,THIN FILTERS CESIUM IODIDE PHOTOCATHODE MICROCHAMVEL PLATES WEDGE AND STRIP ANODE
Fig. 1 . The optical layout of the medium and long wavelength spectrometers sharing different regions of a common detector. The overall layout of the short wavelength spectrometer is similar with the exception that the left and right arrays of mirrors focus on the same region of the detector.
Observations
The diffuse EUV spectrometer was launched April 22, 1986 from White Sands, New Mexico on a Nike boosted Black Brant sounding rocket. The time of launch was set to optimize viewing in the direction of the Earth's shadow cone thereby minimizing geocoronal emission. The Earth's magnetic field traps ionized helium which strongly scatten the bright solar He I1 3046; radiation, and this plasma is shadowed by the earth in the anti-sun direction. A pointing of 1=325O and b=+48' was held for 160 seconds, and then shifted l.SO along the narrow (40') field of view direction and held for another 160 seconds. A comparison of the accumu- lated spectra during the two pointings indicates that no point source of EUV emission contributed to the observed flux. Toward the end of the mission, the instrument made a scan away from the Earth's shadow cone toward the Earth's horizon to test for geocoronal emission and to characterize any airglow component to the spectrum.
The instrument functioned properly during a l l phases of the observation, and the pointings have been confirmed by an independent on-board star camera.
Post-flight calibration of the insaument indicates that the' optical elements of the spectrometer did not shift from their pre-flight alignment.
1 0 0 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 Wavelength (A)
Fig. 2. The resolution (FWHM in 6;) is shown as a function of wavelength. The open symbols indicate measurements made prior to launch, and the solid sym- bols show post-flight measurements. The error bars shown are the 90% confidence intervals to the fits of each measurement, and the curves are parabolas which have been fitted to the data. The scatter about the curves gives a realistic estimate of the errors which arise from trying to mimic a diZfuse source with many small
pencil ray beams.
trum (420 to 6506;) only include 234 seconds of data because the detector used with this spectrometer took longer to stabilize to a constant countrate. The back- ground estimation is shown by the dotted curve in Figure 3. This dotted curve is included to indicate the overall shape of the detector induced background spectrum; it is not the detector background observed during the flight.
Therefore, the solid line histogram in Figure 3 includes the detector background accumulated during the flight in addition to any EUV emission observed.
In addition to the obvious emission line at 5846;, careful inspection of Figure 3 reveals features at 100, 200, 300 and 6356;. To be considered real, however, an emission line must not be an intrinsic feature of the specnometer or due to a transient malfunction of the detector and spectrometer system such as a detector
"hot-spot" or filter tear. Furthermore, a true emission
. .
Preliminary Analysis
The raw data for the three spectrometers are shown in the form of a spectrum in Figure 3. This spectrum
includes all the events observed during the time when o
the instrument was pointed in the anti-solar direction, 1 0 0 - - 2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 and the detectors were operating at constant background WiYelength (L)
levels. The short wavelength spectrum (80 to 2306;) Fig. 3. The raw data from the rocket flight is shown as includes 329 seconds of data. The medium wavelength a spectrum. The dotted line indicates an estimation of spectrum (220 to 4406;) and the long wavelength spec- the detector background.
Table 1 Possible emission lines
(preliminary analysis)
line can not be a consequence of the method used to determine the continuum (or background) level. Detec- tor background images have been accumulated on several occasions before and after the flight, and these background images are used in the following analysis to estimate the (detector induced) continuum so that line emission can be seen in contrast to the continuum. Since the background images are themselves noisy, an alternate method of estimating the background continuum has also been applied. Our analysis utilizes both methods of con- tinuum estimation as any true emission lines should appear regardless of the method used to estimate the continuum Finally, a real detection of an emission line should show a statistically significant rise above the con- tinuum (or detector background) level; this can be deter- mined with ordinary statistical techniques.
The raw spectrum from the flight was fitted with a model composed of the detector background and a number of emission lines. Each line is spread by the instrument resolution at the wavelength of the line and then added to the continuum For an estimate of how well the model fits the data, the Pearson's
x2
statistic was used as described in Larnpton et al. 181. The model was varied by changing the number of lines used to fit the data beginning with no lines and then increasing the number of lines until no further improvement in the fit was noted. The lines which pass all of the above criteria are listed in Table 1.Some notes on Table 1 are in order. The 180A feature is statistically weak, but it is included in Table 1 for completeness. A feature at 610A passes all the cri- teria but it is not included in Table 1 because its wavelength and strength suggest it is the second order ima e of the 3041 emission. Similarly, a feature at 200k has not been included because it is consistent with being the second order image of the 100A feature. The observed counts listed in Column 2 are the total number of counts in the line, and do not include the background counts. The intensities listed in Table 1 have been corrected for atmospheric absorption. Finally, the 304h;
intensity listed in Table 1 is consistent with the upper limit to 304A emission in the anti-sola direction (0.02
Rayleighs) reported by Paresce et al. 191.
Except as noted above, the lines listed in Table 1 pass all the tests to be considered true emission lines.
We have concluded that these lines are not the result of ghost images in the spectrometer system, non- uniformities in the detector response, or malfunctions of the filters or detectors. Furthermore, tests indicated that these lines are from a auly diffuse source and that they are not atmospheric in origin. With the exception of the He I1 304b; emission, none of these lines ap ar to emanate from the Earth's geocorona. The 584remis- sion is clearly the solar radiation resonantly scattering off the neutral helium flowing through the solar system This leaves three possible lines, those at 100, 180 and 6351, which may be produced by the hot interstellar medium.
The limes listed in Table 1 are presented in Figure 4. The non-interstellar lines (304 and 584bl) are indi- cated by squares, and the other lines are marked with
] ! 1 , ' " ' , ' ' 7 T , ~
100 200 300 100 SO0 600 700 lVovclenglll (A)
Fig. 4 . The detected emission lines are indicated: the squares are non-interstellar tines and the circled points are the lines which could be interstellar in origin. The bold liner indicate the upper limits to line radiafion inferred from this observation, and the dashed lines indi- cated the upper limits (and detection at h=114 to 180A) derived from previous observations.
C 1-66 JOURNAL
DE
PHYSIQUEcircles. The regions of the spectrum with no features has been convened to upper limits, conected for atmos- pheric absorption and then plotted with a heavy solid line in Figure 4. Also shown in Figure 4 (dashed curves) is the strength necessary for a single line to pro- duce the broad band detections and upper limits from the Apollo-Soyuz EUV telescope obsmations f2,3/, and the upper limits to line radiation found with the Voyager instrument 151. Note that shortward of 500/(, the upper limits from this observation are the only existing spectral measurements of the diffuse EUV background
Discussion and Conclusions
As can be seen in Figure 4, the short wavelength lines (100 and 180bi) are consistent with the previous observations. The 635bi line is however, above the upper limits set by the Voyager instrument It could be that the 635R intensity is terrestrial in origin and would therefore not be observed by Voyager during its inter- planetary cruise. However, examination of the line intensities as a function of atmospheric path length, and angle from the Earth's shadow suggest that this l i e is not due to atmospheric or geocomnal emission. If the emission is interstellar in origin, it would be subject to strong absorption by neutral interstellar hydrogen and the brightness of the line would be extremely sensitive to the geomeay of the emitting and absorbing regions. It is therefore possible that this is an interstellar emission line which happens to be weaker in the direction observed by Voyager.
The features detected at 100, 180 and 635bi may indeed be'pr+uced by hot gas in the local interstellar medium. In particular, the feature at 635A could be the 0 V resonance emission line at 630i. This feature, along with the upper l i d & derived for the parts of the spectrum with no features, could only be produced by gas near ~ d . ~ ~ i . The feature at l00bl could be produced by several, different highly ionized species such as Si V, Ne W, Ne VIII or Fe JNIII, and the 180A feature could be produced by 0 VI or Fe VIII to Fe M /10,11/.
Further analysis will explore the confidence of these results and implications to models of the hot interstellar medium.
The results presented here are fairly direct evi- dence that the diffuse EUV background is line emission produced by hot interstellar gas. This experiment has produced the first spectrum of the diffuse background
between 80 and 500& and has provided enticing hints to the information available from the EUV background.
Acknowledgements
The spectrometer design was initiated by Chris Martin. We thank Anne Miller for continuing assistance in fabrication assembly and testing, and Aileen Corelli for editorial assistance. This work was supponed by NASA grant NGR 05-003-450 and NGT 05-003-805.
References
1. W. Cash, R.F. Malina, and R. Stem, "An Observation of the Diffuse Soft X-Ray~Extreme Ultraviolet Back- ground," Ap. J. (Letters), 204, L7, 1976.
2. R. Stem, and S. Bowyer, "Apollo-Soyuz Survey of the Extreme-UltravioletlSoft X-Ray Background," Ap.J., 230,755, 1979.
3. F. Paresce, and R. Stem, "The Diffuse Extreme- Ultraviolet Background: Constraints on Hot Coronal Plasma," Ap. J., 247, 89, 1981.
4. F. Paresce, and S. Bowyer, "Upper Limits to the Interstellar Radiation Field Between 775 and 1050
bi
,"Ap. J., 207, 432, 1976.
5. J.B. Holberg, "Fat Ultraviolet Background Observa- tions at High Galactic Latitude: 11. Diffuse Emission," to be published in Ap. J., 1986.
6. R. Kimble, "Observations of the E U V N V Radiation Fields," Doctoral Thesis, University of California, Berke- ley, 1983.
7. S. Labov, S. Bowyer, and C. Martin, "A spectrometer to measure the diffuse, astronomical extreme ultraviolet background," Proc. SPIE, 627, 379, 1986.
8. M. Lampton, B. Margon, and S. Bowyer, "Parameter Estimation in X-Ray Astronomy," Ap. J., 208, 177, 1976.
9. F. Paresce, H. Fahr, and G. Lay, "A Search for Inter- planetary He 11, 304R ~missio"," J. Geo. Res., 86, 10,038, 1981.
10. R. Mewe, E. Gronenschild, and G. van den Oord,
"Calculated X-radiation from optically thin plasmas. V."
Ast. Ap. Supl. 62, 197, 1985.
11. R. Stem, E. Wang, and S. Bowyer, "Extreme Ultra- violet Spectrum of a Hot Interstellar Plasma," Ap. J.
sup., 37, 195, 1978.
