3.3-~m Spectroscopywith IRSPEC · 1.10-1.18 or 3.1-3.5~m) where at mospheric transmission is not...
Transcript of 3.3-~m Spectroscopywith IRSPEC · 1.10-1.18 or 3.1-3.5~m) where at mospheric transmission is not...
3.3-~m Spectroscopy with IRSPECT. LE BERTRE, ESO
3 .~;;50;---~-3:;-.-;;:20:;;;0:----'~-:3;-.~25;';;"0-~---;3;-.3:;;0:;;"0---'---;;3-:.3:::5-:::-0-~--;;3-.•~00:---~-3-.•.l.J50
VI AVE L ENG T H (11 ml
Figure 1: Raw spectrum ofa K2/11 star. The strang tel/uric feature at 3.32 ~Im, due to Methane,is indicated by an arrow.
IRSPEC is a cooled grating spectrometer in service at ESO since October 1986. It is equipped with a lineararray of 32 InSb detectors and with twoback-to-back-mounted gratings; in itspresent stage, it permits spectroscopyin the 1-5 ~lm region at a spectral resolution (R =)J A.) of 1,000 to 2,000, depending on grating, order andwavelength. A description of it and areport on its performances can be foundin Moorwood et al. (1986).
Although, in principle, data can beacquired at any wavelength between 1and 5 ~m, astronomical observations insome spectral regions (for instance,1.35-1.45 ~m or 2.5-3.0 ~m) are imPossible due to the large atmosphericextinction. There are regions (e. g.,1.10-1.18 or 3.1-3.5 ~m) where atmospheric transmission is not zero, butstill poor. There is great interest in performing spectroscopy in the 3.0-3.5 ~m
range. Such observations are difficult,not only due to the bad atmospherictransmission, but also due to the thermal background which rises in thisrange. Not much can be done by regularusers on the thermal background issueand, hereafter, only the first source ofdifficulties will be addressed.
Figure 1 shows the raw (3.15 3.45 ~m) spectrum of a K2111 star. Thisspectrum was obtained in May 1988during a good photometric night; relative humidity at ground level was-10%. It has been acquired by stepping the grating at three different positions; the spectrum segments are contiguous and do not overlap. Spectral resolution is - 1,100. At this resolutionand in this range, a K2111 star is notexpected to present any strong spectralfeature and, therefore, what is presented in Figure 1 is basically the atmospheric transmission. One notesmany absorption features, some ofwhich are due to several overlappinglines and could be resolved at higherresolution. A feature at 3.318 ~m is particularly deep and strongly affects observations even at low resolution. A lowresolution spectrum (R - 80) of the atmospheric extinction can be found inMartin (1987); this spectrum, obtainedwith a circular variable filter (CVF),shows a peak around 3.3 ~m. It is interesting to study this kind of spectrumbefore considering observations in the3.0-3.5 ~m range. As several atmospheric lines of different intensities (someof which are even saturated) affect thesame spectral element, it is of primeimportance to observe scientific target
and comparison star at airmasses asclose as possible; using the comparisonstar spectrum as a template allows, inprinciple by simple division, to correctthe target spectrum for all its atmospheric features. However, experienceshows that even during good nights,atmospheric absorption in the3.0-3.5 ~m range is not always stablewith time. Therefore, it appears advisable to select a comparison star asnearby in the sky as possible from theobject, and to observe it before andafter in order to control the stability ofthe atmospheric transmission. Also, toavoid variation in airmass during exposure, one should observe preferentiallyaround the meridian. It is only by carefully applying such guidelines that theobserver might be able to cancel atmospheric features in its spectra. This advice is relevant to medium resolutionspectroscopy with IRSPEC and, a fortiori, to low resolution spectroscopywith CVF's.
Among the most exciting observations to perform in the 3.0-3.5 ~m region are those designed to study the socalled unidentified infrared emissionfeatures. The most prominent one iscentred at 3.3 ~lm and has long beenobserved without being satisfactorilyexplained. Leger and Puget (1984) haveproposed a consistent physical interpretation of it in terms of polycyclic aromatic hydrocarbon (PAH) molecules.Another feature, generally less intense,is observed at 3.4 ~m. These two
features are, in general, correlated withothers found at 6.2, 7.7, 8.6 and11.3 ~m which may, therefore, have asimilar origin. However, till now, most ofthe astrophysical observations havebeen done at low resolution (R - 80)with CVF's, and do not permit profilestudy of these features. With the availability of grating spectrographs likeIRSPEC, physical studies based onmedium resolution spectroscopy arefeasible; also, now, comparison withlaboratory data should be more meaningful in identifying the carriers. For instance, working at aresolution of 400,de Muizon et al. (1986) have observedstructures in the 3.3-~lm and in the 3.4~m features and have discovered threenew features at 3.46, 3.51 and 3.56 ~m;
accurate wavelengths, profiles and relative intensities of these different featuresgive information on particle size, chemical composition, ionization state, etc. ofthe carriers.
To illustrate more specifically the potential offered by IRSPEC, a spectrum inthe range 3.15-3.55 ~m of Hen 1044 ispresented by a WC star. Its progenitor isprobably an intermediate mass (1 MG <M < 9 MG) star evolving from theAsymptotic Giant Branch (AGB) to thewhite dwarf stage. The high excitation ofthe nebula is clear from its H-bandspectrum where Brackett lines up to(20-4) can be seen. Cohen et al. (1985)studied its IRAS low resolution spectrum (LRS); they show that, in additionto features at 7.7 and 11.3 ~m, this
ESO 36mMay 30/31 1988
55
Hen 1044ESO 36mMay 30/31,1988
3.~ 3.~ 3.~ 3.~ 3.~ 3._ 3.~ 3.~
W AVE L ENG T H '~m\
Figure 2: 3.15-3.50 pm spectrum of Hen 1044. The expected position of Pf bat 3.296 tim isindicated by an arrow.
CVF's have not been confirmed later;the reason may simply be a rapid variation of the atmospheric absorptionaround 3.3 ~lm.
In conclusion, spectroscopic observations in the range 3.0-3.5 /-lm arefeasible at La Silla using the 3.6-m telescope equipped with IRSPEC andpromising prospects are offered by theutilization of this instrument. However, acareful preparation of the observationsis required to get useful data; the observing planning should not be improvised at the telescope. Such a preparative work is surely time consuming andpainful, but, at the end, it may make thedifference between a successful mission and an unsuccessful one.
W AVE L ENG T H !Ilm)
Figure 3: 3.15-3.50 ,/Im spectrum of Hen 1379. The "emission fine" at 3.32 1,m is an artifactdue to atmospheric Methane (see Figure 1).
STAFF MOVEMENTSArrivals
Europe:
ALBERTH, Manuela (0), AdministrativeClerk Purchasing
BECKERS, Jacques (USA), Experimental Astrophysicist
JANSSEN, Edmund (NL), OraughtsmanOOSTERLOO, Thomas (NL), Fellow
3.~3.~3.400
PIENEMAN, Hendrik (NL), AccountingAssistant
PIRENNE, Benoit (B), Fellow (ScienceArchive Software Specialist)
PRIEUR, Jean-Louis (F), FellowVAN RIJN, Gunilla (NL), Administrative
AssistantWALLANOER, Anders (S), Software En
gineerWIELANO, Gerd (0), Procurement
Officer
3.~
ReferencesCohen, M., Tielens, A. G. G. M., Allamandola,
L.J.: 1985, Astrophys. J. 299, L 93.Leger, A., Puget, J.-L.: 1984, Astron. Astro
phys. 137, L5.Martin, W.: 1987, Astron. Astrophys. 182,
290.Moorwood, A., Biereichei, P., Finger, G., U
zon, J.-L., Meyer, M., Nees, W., Paureau,J.: 1986, The Messenger, 44,19.
de Muizon, M., Geballe, T. R., d'Hendecourt,L.B., Baas, F.: 1986, Astrophys. J. 306,L105.
3.~3.~3.200
Hen 1379ESO 36mMay 30/31.1988
3.150
spheric features are weil cancelled, except the strong telluric absorption at3.318 I.lm which varied during the observations in such a way that an "emissionline" appears around 3.3 /-lm. At CVFresolution, such an effect may mimic theunidentified 3.3-/-lm emission; with theresolution of IRSPEC, there is no ambiguity on its telluric origin: the 3.3-~lm
feature is clearly absent from the spectrum of Hen 1379. Some reported detections of the 3.3-/-lm feature with
source presents a plateau of emissionextending from 11.3 to 13.0 /-lm. Theydiscovered this new feature in the LRS'sof 20 IRAS sources which exhibit emission at 7.7 and 11.3 /-lm and attributed italso to PAH's.
The spectrum presented in Figure 2has been corrected for atmospheric extinction as explained above; no othertreatment of the data (smoothing, etc.)has been performed and its resolution isexactly the one given by IRSPEC in thiswavelength range (i. e., R -1100). Asthe object is bright (L - 4.3), the noise inthe final spectrum is mainly due to imperfect cancellation of atmosphericfeatures. The 3.3-/-lm emission is clearlyseen in the Hen 1044 spectrum; thefeature is resolved and presentsasymmetrie wings. Adjusting it with agaussian profile (which is somewhatequivalent to degrading the resolution ofthe spectrum), one finds a centralwavelength of 3.292 J.lm and width of.044 /-lm. However, as the contributionof the Pfund Ö hydrogen line (9-5) at3.296 /-lm might not be completely negligible, a careful analysis is needed; thisline is expected to affect at most twocontiguous pixels. An interestingcharacteristic of this spectrum is thetotal absence of feature at 3.4 /-lm andlongwards. The clear absence of emission at 3.4 /-lm suggests that thefeatures at 7.7 and 11.3 /-lm and theplateau at - 12 /-lm are not correlatedwith it. Such recognition of correlation(or absence of correlation) betweenfeatures, in combination with laboratorystudies, will allow to constrain the identification of the different PAH's existingin astrophysical environments (see, forinstance, de Muizon et al., 1986). Thecomplete infrared spectrum as weil as adiscussion of all the data acquired onthis interesting object will be presentedelsewhere.
In Figure 3, the spectrum of Hen 1379(another post-AGB object studied bythe author) is presented. Most atmo-
56