f

Indian Journal of Radio & Space Physics Vol. 28, August 1999, pp. 191-201

Space-borne infrared spectro-photometer for astronomy*

S KGhosh Space Physics Group, Tata Institute of Fundamental Research, CoJaba, Mumbai (Bombay) 400 005

Received 16 November ,1998; revised 30 March 1999; accepted 4 May 1999

A simple and reliable satellite-borne infrared spectro-photometer (IRSP) for astronomical applications is proposed, which can be realized in the near future using the available technological knowhow within the country. While this instrument 's demands for cryogenic and oriented platform-pointing systems are kept modest (within the anticipated reach of ISRO capabilities), it can still address a wide variety of interesting and contemporary astrophysical problems.

1 Introduction

With the regular and successful launches and operations of Indian Space Research Organization'S (ISRO) 'augmented satellite launch vehicle (AS LV)' initially, and more recently 'polar satellite launch vehicle (PSL V)' rockets and their numerous payloads, it is now time to consolidate the technical advancements with fundamental research in pure sciences for the near future. Already several attempts in this direction have been made in the recent past, which have seen spectacu lar successes, e.g. gamma ray burst (GRB) experiment using the ASL V (Ref. I) and Indian x-ray astronomy experiment (IXAE) using the PSLV (Refs 2, 3). Although the GRB experiment was a modest beginning leading to new science results4

, the IXAE is more sophisticated and uses a pointed platform. The IXAE continues to do world class astronomy even toda/ ·6, although it has been sharing the satellite platform with a host of other application / engineeri ng-oriented payloads to econo­mize the resources.

The above provides ample proof that the technical and scientific knowhow in the country has reached a maturity where the astronomy community as well as ISRO anticipates an ent ire space mission fully devoted to astronomy, thereby avoiding several hurdles which were faced due to sub-optimal features (e.g. choice of orbit) of earlier shared missions. In addition, opportunities for small and simple

*Based on a presentation made at the Indian Space Research Organi zation Head Quarters, Bangalore, during a meeting on ' Indian Multi -wavelength Astronomical Satellite' during 30-31 luly 1996.

astronomy experiments as piggyback payloads to non­astronomy missions may become available regularly.

In the aforesaid backdrop, ISRO has already initiated action involving scientists interested in designing and developing potential space astronomy experiments for a future Indian multi-wavelength astronomy satellite (!MAS). The concept of one such experiment - a near infrared spectro-photometer (IRSP) for astronomy, is described here, which is not only technically feasible, but also has potential to study a variety of problems of contemporary interest. Even if IRSP does not form a part of !MAS, it has great potential as a piggyback experiment onboard some other Indian mission.

2 Design philosophy

From the point of view of the instrument builder, the following design philosophy has been adopted : Within the practical limitations constrained by the available technology and knowhow, how best the infrared spectro-photometer (IRSP) can be designed and developed within a well defined time frame (- 3-5 years).

The two most crucial constraints are: (i) the cryogenic capabilities which decide the lowest achievable temperature for the infrared detector; and (i i) orientation and pointing capabilities of the stabilized platform on which the present instrument will be mounted.

The cooling capability decides the type of the infrared detectors that can be employed and hence, in tum, the longest possible usable infrared wavelength. In addition , the dark current of the detector crucially depends on this temperature, thereby limiting the

192 INDIAN J RADIO & SPACE PHYS, AUGUST 1999

maximum usable on-target integration time which, in tum, fixes the achievable sensitivity. From the point of view of scientific capabilities, longer infrared wavelengths are more desirable. The characteristics of the earth' s atmospheric transmission is such that, longer the infrared wavelength, less accessible is the waveband from the terrestrial te lescopes, thereby, less known are the specific astrophysical phenomena I problems re levant to these wavebands. At longer infrared wavelengths, the utilization of the space­borne platform is also more effective as the thermal background emission gets worse at longer wavelengths .

The orientation and pointing capability decides the smallest effective fi e ld of view usable ·for astronomical imaging or spectroscopic measurements . The fi e ld of view is an important factor for the photon background on the detector element (pixel) which can limit the sensi tivity of the instrument for pointlike as well as ex tended astronomical sources .

Hence, it is clear that the cryogenic and orientation capabilities ultimate ly dec ide the nature of ast rophysical problems that can be addressed by the space as tronomjcal instrumentation. For the present, both these cruc ial parameters are taken to be variable and their effects on the instrument design as well as ach ievable sc ience from a near infrared spectro­photometer are di scussed in detail. The orientation capability has been assumed to be at least better than 0 .1 degree, since (i) such capability has already been demonstrated by ISRO in its IRS-P3 mission which currently incorporates the Indian x-ray astronomy experiment (lXAE); and (ii) also poorer orientation makes the science return from an infrared instrument uncommensurate with the anticipated investments in technical efforts and finances for developing the instrument.

3 Cryogenics

Cryogenic capability is the most important issue having bearing on the infrared wavelength in which astronomically meaningful measurements can be made. In any cryogenic system, the achieved temperature obviously depends on the total heat load. The heat load is contributed by parasitic heating, spacecraft environment (which very strongly depends on the orbit parameters) and the local dissipation within the detectorlinstrument. \

The most optimal thermoelectric cooling can reach as low as - 180 K with a multistage unit. However, the cooling efficiency drops drastically as one cascades stages. Passive radiative cooling is quite effect ive in very high orbits (in geostationary orbits, one may even reach -80 K), but can be very restrictive in lower orbits (on ly up to - 150 K) . Open cycle solid cryogen (N2) is being used in the near infrared camera and mu lti-object spectrometer (NICMOS) . instrument onboard Hubble space telescope (HST) with an estimated life time of about 5 yr, keeping the detector at 58 K. For achieving even lower temperatures, mechanically more complex refrigerators are needed, which have several moving parts . Closed cycle refrigeration (Stirling/Vuilleumier IGifford-McMahon) system can reach up to 10K depending on the compressor capacity vis-a-vis heat 10ad7.8 . To give a feel for the cryogenic systems used in various recent space missions7

, Table I lists relevant details.

For the present study, it has been assumed that a small mass comprising the ncar infrared (NIR) detector and its housing is cooled to 77 K (perhaps using a mechanical Stirling cooler aided by efficient passive cooling) and the optical components are at 0- 160.K (using only passive cooling). However, to keep the study useful, even if assumed situation is not achieved in practice, the peliormance penalties for higher temperatures have been quantified.

----------------Table I-Examples of spacecraft and instruments employing cryogenic technology 7

Instrument Temperature Required Heat load Type

K lifetime mWatt

IRIS 200 5 years 500 Radiator

WFPC 178 3 years 60 Thermoelectric cooler

SISEX 120 2 weeks \000 Stirling refrigerator

GRS 92 8 months 50 Solid cryogen

NIMS 75 > 3 years 50 Radiator

ATMOS 75 2 weeks 500 Stirling refrigerator

SFHE 1.5 I week Liquid cryogen

GHOSH: SPACE-BORNE INFRARED SPECfRO-PHOTOMETER FOR ASTRONOMY 193

4 Photon background

There are two components to the photon background at NIR wavelengths as expected at spacecraft altitudes: (i) the thermal emission from the warm telescope optics (light gathering system) itself, and (ii) the zodiacal background originating from the solar system. The former is a modified Planck function sensiti ve to the temperature of the optical components as we1l as their emissivity properties at relevant wavelengths, whereas the zodiacal .background comprises the scattered sunlight and the thermal emiss ion of the interplanetary dust. From the cosmic background explorer (COB E) rruss lOn measurements9

, it has been found that the total zodiacal background can be represented as

Z (A, To)" = ( 1.01 x I09 / A L69) +6 x IO-8 B (A., To) h - I - 2 _ °1 - I ( I) P otons s cm Ilm sr ...

at ecliptic latitudes of - 45° . The first term in the RHS of Eq. ( \) represents the scattered sunlight and the second term the thermal component. The parameter B is the Pl anck function and To the temperature of the interpl anetary dust (typica1ly - 265 K).

In order to apprec iate the ro les of thermal emiss ion from the te lescope vis-a-vis zodi acal background, the photon counts, as a functi on of wavelength, are presented in Fig. I for several temperatures of the telescope 10 .

7

6 10 ME ER TEL SCOPE T 80, 1 O,ond

ombdo 20 PI •• I. lodlocol IOn' lOll IEml"""i~ lIock9'0

4

J

2

/

o / -1

o 2 J 4

The advantages of the space-borne NIR instrument over the ground-based one (in addition to the accessibility of the wavelength from above the earth's atmosphere) can be quantified in terms of the photon background at a wavelength in an atmospheric window as expected at the best telescope site on the earth's surface". The latter is presented in Fig. 2. Here, in addition to the thermal emission, the background consists of (i ) airglow due to the vibrational-rotational spectrum of the radicals/ molecules (OH and O2), (ii) the nightglow continuum and (iii) the polar aurora. The airglow originates in the atmosphere at about 90 kIn above the earth ' s surface, and hence cannot be avoided by choosing a higher mountain site for astronomical measurements . Other problem about airglow is that their emission is variable with time, changing by factors of over 2 in less than half an hour II.

5 Survey of infr" red astronomical mission

If our proposed NIR spectro-photometer has to be competiti ve intemationa1l y, it has to be exclus ive enough with minimum overl ap with any other space mi ssions. Since the atmospheric windows, viz., J, H and K bands, are access ible from large terrestrial telescopes equipped with far more sophisticated focal plane instruments, new science is expected mainly from the wavelengths in between rhese bands. However, IRSP should cover the entire near infrared range ava ilable (depending on the capabilities of the

.. ~ 120 K

20 K V ~ .....

...... ' 100 K n. ,/ /' l ........

V /' , .. 80 K

" / / ~ ~

~

l 00"" ( ~ 9AC~ ROUND

V~ ~

6 7 8 9 10 11

WAVELENGTH, ).lm

Fig. I-Comparison of zodiacal background with the thermal background (after Tho mpson 10)

194 INDIAN J RADIO & SPACE PHYS, AUGUST 1999

10'

";"

'IJ~v.'" E 10' :1. ... .:-.,

10' 0 0 til

0 t- to' ell '-' ~

E 10' , til ?-

>=' 10' f-c ....... CI)

1000 Z r..u

~ 100

10

1-----,,-

2 3 " 5

WA VELENGTH, ~m

Fi g. 2- Typical pholon background spectrum at Mauna Kea (4200 m above mean sea level; after McLean I I)

detector used), including the J, Hand K bands. The capabilities of IRSP in the J, Hand K bands will lead to: (i) improved in situ calibration of the instrument (since enormous database exists at these wavebands) and (ii ) an edge over the terrestrial experiments for some spec ific astrophysical problems sensitive to the photon background and background fluctuations in these bands (at terrestri al sites, the maj or background in JIHIK bands is due to the airglow which is absent for sate llite-borne experiments).

In what follows, we consider ~he infrared missions havi ng any coverage in the 1-1000 11m wavelength band . The past and proposed missions include : (a) Infrared astronomy satellite ' 2 [IRAS ; all sky survey at 12, 25, 60 and 100 I-lm with limited angular resolution (5 arc min) ; NASA / Great Britain / The Netherlands; 1983], (b) Cosmic background explorer'3 [COBE; all sky maps at NIR to millimetre wavebands (3-1000 11m) with wide ( 1°_7°) beams; NASA; 1990], (c) Infrared telescope in space l4 [lRTS ; near, mid and far infrared instruments for photometry and medium resolution spectroscopy ( 1-12 I-lm and loo-IOOOl1m); Japan; 1995], (d) Infrared space observatory'5 [ISO; photometry , imaging, polarimetry and medium-to­high resolution . spectroscopy (2.5-200 I-lm); ESA; 1995], (e) Near infrared camera and multi-object

spectrometer on board Hubble space telescope l6

[NICMOS; imaging and grism based medium resolution spectroscopic imaging (1-2.5 I-lm) ; NASA; 1998], (f) Space infrared telescope facilit/ 7 [SIRTF; imaging and high resolution spectroscopy (3-300 ~!m); NASA ; 2002] and (g) Far infrared and sub­millimetre telescope 's [FIRST; high resolution spectroscopy (50-400 11m); ESA; 2007] .

Our present proposal of IRSP in the near infrared has wavelength overlap with only the IRTS and NICMOS missions. These two are considered here with little more details.

The Japanese IRTS was a 35-day mission with 15 em primary mirror (8'x8' field of view). The NIR spectrometer covered 1.4-4.0 11m with a spectral resolution of !1'A = 0.12 I-lm. Two one-dimensional arrays of Indium Antimonide (lnSb) of 2x12 pixels were cooled to 1.8 K. Their science goals included the study of zodiacal light, diffuse Galactic light, extragalactic background light and point sources brighter than 7th magnitude in the K (2.2 I-lm) band . The IRTS was a survey-type mission with no pointed observations of specific sources. That explains their large field of view (FOY) . Hence, our science goals will not clash with those of IRTS .

f

GHOSH : SPACE-BORNE INFRARED SPECTRO-PHOTOMETER FOR ASTRONOMY 195

Although NICMOS is a true competitor to IRSP, but being very sophisticated, its emphasis is on much fainter samples (2.5 m HST primary mirror). In addition, it is extremely oversubscribed by the entire international community. Also, NICMOS use is mutually exclusive to other focal plane instruments currently available in HST (which are all in the optical / UV wavebands).

At wavelengths longer than 2.5 ~m, IRSP would have had overlap with ISO and SIRTF, but due to cryogenic restrictions, IRSP is planned for 0.9-2.5 ~m waveband only.

6 Infrared spectro-photometer

In keeping with the technical feasibility and the international scene, the infrared spectro-photometer (IRSP) is proposed to cover the entire wavelength range of 0.9-2.5 ~m. We define and concentrate on /J (1.0-1.2 11m, i.e. the gap between I and 1), lH (1.3-1.4 11m) and H K ( 1.8-2.0 ~m) photometric bands in addition to the atmospheric windows 1 (1.15-1.35 ~m) , H (1.45-1.8 ~m) and K (2.0-2.4 ~m) . In the spectroscopic mode also, the entire wavelength range from 0.9 to 2.5 11m is covered to make calibrations of various instruments and their comparisons with grouod-based measurements simpler.

Since Cassegrain optics offers a very compact geometry, it is favoured in satellite configurations, e.g. the !MAS proposal for an UV experiment by Pati and Rao l9

. Figure 3 shows, schematically, the proposed position of IRSP at the Cassegrain focus of the telescope. The primary aperture of this telescope has been assumed to be of a very modest size of 30 cm (diameter) for computing various details , e.g. background, sensitivity , etc . The IRSP is proposed to consist of a photometer with six broad bands (covering IJ, lH. HK, 1, Hand K bands) and a medium resolution (R - 100) spectrometer covering the wavelength range 0.9-2.5 ~m.

There will be no moving part in IRSP to achieve high degree of reliability . All the bands of the photometer wi II be arranged to view the sky simultaneously (say position A in the sky). The spectrometer will cover the entire wavelength range 0.9-2 .5 ~m simultaneously (say the position B of the sky). Since the positions on the sky A and B will be offset (by a few arc minutes) for a particular source, the photometric and spectroscopic modes are mutually exclusive.

~ 1,\OPliC AXIS

I IN~~yESNl I ~ SECONDARY 1..--- - ........ I ~ I \l MIRROR

I , I I I , , I I I I I I I

I I

I I I I

I

, I

I

j I

I I

I

I I

I

I' . I'

PRIMAR Y MIRROR

-.--+-------, NEAR NEAR

INFRARED INFRARED

SPECTROMETER PHOTOMETER

Fig. 3-Schematic diagram of the proposed layout of infrared spectro-photomeler at the Cassegrain focus of the telescope

The state-of-the-art near IR detectors (astronomy grade) are either made of Indium Antimonide (InSb) for 1-5 ~m wavelength coverage or Mercury­Cadmium-Telluride (HgCdTe) for 1-2.5 ~m coverage. The InSb arrays are currently manufactured by Santa Barbara Research Center (USA), while the HgCdTe arrays are manufactured by Rockwell International Science Center (USA). Whereas the optimum operating temperature for the HgCdTe arrays is about 70-80 K, the cryogenic requirement for InSb arrays is more severe (optimum operating temperature - 35 K) . Since our interest is restricted to 2.5 ~m, HgCdTe arrays are preferred . A 256x256 format HgCdTe array (Rockwell) with 40 ~m square pixel, has

196 INDIAN J RADIO & SPACE PHYS, AUGUST 1999

recently been used in space for the NICMOS instrument onboard the Hubble space telescope (HST) . Quantum efficiency of this type of arrays is 40-60 % throughout the responsive wavelength range and a read-noise of 30-50 e - is achieved routinely. The well capacity of 5x 105 e - is quite common. The dark current varies between 10.0 and 1.0 (e-/s/pixel) corresponding to array temperature of 85-60 K.

For IRSP, we propose to use only a one­dimensional array with 256 elements for the spectroscopic part and much smaller arrays for the photometric segments. In order to estimate the capabilities of IRSP, the following assumptions have been made, which are rather conservative:

(i) The achieved read-noise of the detec tor and its drive electronics system is - 100 e-.

(ii) The well capacity is IxlO5 e-. (iii ) The detector is maintained at a temperature of

77 K.

In addition , the pOll1tll1g jitter of the telescope IS

assumed to be - 0.3 arc sec.

6.1 Background

Here, we estimate the NIR background for the space-borne IRSP and demonstrate its superi ority over the best terrestrial site (in addition to the accessib il ity aspect of the particular wavelengths) .

We use the experience9 of the HST and sca le the co llect ing area (30 cm dia primary feeding for IRSP), pixel solid angie ( 1.2 arc sec dia, circular). The

background es timate is 10 photon/s at 2.1 Jlm for a bandwidth of !1A = 0.6 Jlm. The background estimated 0 11 the bas is of SIRTF miss ion documents (again sca led to IRSP configuration) leads to a number - 4 photons/s in the same waveband. Both the estimates are reasonably close. One may compare these

numbers with the estimated background at Mauna Kea" (the best terrestrial site for infrared astronomy, -4200 m above the sea level) for IRSP-like instrument to be -1600 photons/so The big advantage that IRSP will have over terrestrial experiments (at nearby wavelengths) is, thus, obvious .

6.2 Sensitivity

The expected sensitivity of IRSP in the photometric mode (R = N!1A =: 3.5) is presented in Table 2 for IRSP system with 30 cm dia primary mirror; 1.2 arc sec pixel, HgCdTe at 77 K as detector (with 100 e- read-noise) and quantum efficiency of 0.4 . The total effective transmission of the entire optics (telescope and fore optics) has been taken as 0.15 . The anticipated degraded sensitivities, in case the read-noise is larger by a factor of 2 (i .e. 200 e -) , are also listed. It must be noted that although the sensitivities are being quoted for the familiar bands lIHIK at the terrestrial atmospheric windows, IRSP photometer will have additional bands covering wavelengths in between these bands. The lIHIK bands are being used here for ease of comprehending the implications of the sensitivity numbers due to the familiarity of magnitudes in these bands.

Table 2 also gives the break-up of various limiting factors to the sensitivity of IRSP in the K band for a given integration time. Then the same is repeated for different integration times. F inally, the systematic effects make further increase in the integration time fru itless.

The limiting magnitudes for the J and H bands are listed in Table 3.

Using the stellar properties, the above sensitivites can be transl ated to the di stances up to which different types of stars can be studied with a minimum stipulated signal-to-noi se ratio . These numbers are presented in Table 4 .

Table 2-Expected sensi tivity in the K band

Integration Photon Dark Read- Total· K mag

background current noise noise 3a

e e e e

1.2 10 100 -100 11.3

(200) ( 10.5)

10 12 100 100 -101 13.8

(200) (13.0)

100 120 1000 100 -1 05 16. 2

1000 1200 10000 100 -146 18.3

* The numbers in parentheses refer to a read noise of 200 e-

GHOSH : SPACE-BORNE INFRARED SPECTRO-PHOTOMETER FOR ASTRONOMY 197

6.3 Pointing jitter penalty

In the event of degraded perfonnance of the spacecraft in keeping the telescope optical system / IRSP pointed on the sky (i.e . poorer than the assumed value of 0.3 arc sec r.m. s.), the pixel size in IRSP needs to be increased to make meaningful observations. The pixel size must be several times the r.m .s. pointing j itter. This will increase the thennal and sky background leading to a penalty in the achieved sensiti vity for point-like sources (stars). Thi s factor, termed as ' pixel size penalty ', has been computed and listed in Table 5. It may be noted that, for any fi e ld of view, there is a limit on the largest integrat ion time possible because of the finite well size of the detector (leading to saturation; denoted by an asteri sk in relevant entries of Table 5) .

6.4 Detector temperature penalty

If the cryogenic temperature achieved for the infrared detector is hi gher than 77 K (which is assumed for all the above sensitivity estimations), there will be severe performance degradation of

Table 3-Limiting magnitudes in J, Hand K bands

Magnitudes for band

s J H K

12.6 12.0 11.3

10

100

1000

IS . I

17.5

19.6

14.5 13.8

16.9 16.2

19.0 18.3

Table 4-Absolute magnitudes of late type stars

Ste ll ar Super-giant Giant Main type (l a) (III ) sequence (V)

Mv, V-K Mv,V-K Mv, V-K

KO -7. 5,2.2 +0.5,2.3 +5.9 , 1.8

K5 -7.5,3.7 -0.2.3.7 +7.3,2.8

M2 -7.0,4. 1 -0.6,4 .1 +10.0, 4.3

M5 0.8 ,6. 1 + 11.8,5.2

Table 5- Pixel size penal ty due to poor pointing

Tint Achieved sensi tivit y (K magnitude) for pixel size (9)

1.2" 5" 20" 1.3' 5'

11.3 11.3 11.2 I 1.1 10.1* . 10 13.8 13.8 13.6 12.8*

100 16.2 16. 1 15.5*

1000 18.3 18.0 17.5*

( 10")

* Saturation due to detector well capacity limit

IRSP. This aspect has been quantified here. A higher detector temperature leads to much higher dark current. First, this will severely limit the largest integration time and, in addition, the fluctuations in dark current will constitute a noise component which may dominate over all other . noises, thereby degrading the sensitivity drastically. The dark current being exponentially dependent on the reciprocal of the detector temperature [propoItional to e (-Eg/kT) ;

where Eg is the band gap for the detector material], the perfonnance degradation can be very drastic. The band gaps for InSb and HgCdTe are 0 .23 eV and 0 .5 eV, respecti ve ly. The loss in the performance is tabulated for both these detector material s in Table 6.

6.5 Penalty due to the spectroscopic resolution

So far the IRSP sensitivity has been di scussed for the photometric mode [R = IJ!J.'A = 3.5] . However, in the spectroscopy mode, since the photons from the astrophysical source get further di stributed over multiple resolution elements (pixel s of the one­dimensional array), the sensitivity (in tenns of stellar magnitude) reduces.

The worst case loss in sensitivity translated to magnitude scale, for a flat spectral energy distribution, has been estimated. The resulting numbers have been displayed in Table 7. It may be noted that the real loss may be milder in case the detector is operating in the background limited mode (referred to in the lite rature as BLIP : Backgrou nd Limited Pefonnance).

7 Attractive science with IRSP

Fortupate ly, there are several important atomic/ nebular/molecular lines of astrophysical interest, which fall in between the successive terrestrial

Table 6-Deteetor temperature penalty

Temperature (T) Noise (T) I Noise (1;,) Magnitude loss

For Hg Cd Te (To = 77 K)

77 1.0 0.0

85 35 3.8

90 224 5.9

100 6xlO3 9.3

For InSb (To = 50 K)

50 1.0 0.0

60 84 4.8

70 2xlO3 8.3

80 2x lO4 10.8

198 INDIAN 1 RADIO & SPACE PHYS, AUGUST 1999

Table 7-Penalty due to spectroscopic resolution

Resolution (R)

10

25

100

Magnitude loss (MJ

\.I

2.2

3.6

atmospheric windows (and, hence, not accessible from the ground-based telescopes). These lines can be studied by IRSP very fruitfully. These lines are listed i.n Table 8 alongwith the typical spectral. resolution required to detect them well above the anticipated continuua.

The expected strengths of some of the atomic and

molecula, hydrogen lines under typical astrophysical

conditions (Te=Ne= 104) are given in Table 9. These

have been normalized to the strength of the most

commonly studied line from the same species.

The astrophysical importance of these lines, which

will be accessi ble to IRSP but not to terrestrial telescopes, is highlighted next. The recombination lines of hydrogen , viz., Paschen-a, Brackett-o, 4-3,

etc. will be very useful in understanding the structure

of Galactic H II regions and radiative transfer through

them. The [ Si VI ] line measurements are crucial for

studying winds of hot stars and structures of their

corona. The line emission from molecular Hydrogen

will provide very significant clues to understand

atmospheres of cool stars in general, and M dwarfs in

particular. Since the theoretical models of outer

envelopes of cool stars are very unsatisfac tory at the

presen t time, data from these H2 lines will help a great deal in settling various issues related to their structure and assu mptions about turbulence: Our

knowledge about the origin and location where the

ste ll ar winds form is likely to improve from the above

NIR spectroscopic measurements . Finally, the lines from the other molecular species, viz., H20 , HCOl . C2

and CO. are ex tremely useful for various studies of

astrochemistry in the general Galactic interstellar medium, star forming regions as well as local interste llar medium. The branches which will particularly. benefit are formation of these molecules. the ir abundances and excitation mechani sms of these lines (e.g. collisional versus

Table 8-Resolution requirements for astrophysically important lines not accessible from terrestrial telescopes

Wavelength Line identifier Resolution needed

Jlm 1.87 Paschen-a -100

1.94 Brackett-o -100

1.96 [SiVlj -100 1.45 H2O -\0

1.8 HC02/C2 -25

2.4 CO -10

Table 9-Expected strengths of atomic and molecular lines of hydrogen which are not accessible from terrestrial telescopes

Transition

Atomic Hydrogen

(Nu-N,)

4-3

8-4

9-4

Molecular Hydrogen

1-05(3)

1-05(4)

1-05(5 )

I -OQ(I)

I - OQ(2)

I-OQ(J)

Wavelength Normalized line Jlm strength

I (A.) / I (HII)

1.87 0.33

1.95 0.02

1.82 0.01

I (A.) I I (1-0 S (1)

IOOOK---4000K

I. 96 0.5 1----- 1.45

1.89 0.08-----0.47

1.84

2.40

2.4 1

2.42

0. 10- ·--- 1.2 1

1.05-----0.57

0.30-----0. 21

0.70-----0.70

radiative). Several other long term sc ientific

programmes which can be taken up using IRSP are

di scussed as follows.

7.1 Late type giants / main sequence

KIM Giants:

Even with a 5" pixel size, I s integration time and

R = 100, a 30' detect ion limit leads to a di stance modulus of 12 magn itudes or 2.5 kpc. This implies

that a large sample of K III (3x 107) and M III (2x 106

)

stars in the solar neighbourhood can be studied spectroscopically . Spectral lines, if present, will be detectable at comfortable signal-to-noise level. Even main sequence (MS) stars K5-M2 can be studied

spectroscopically up to a distance of 34 pc (the number of K V stars - 1500; the number of M V stars

- 7000).

GHOSH: SPACE-BORNE INFRARED SPECfRO-PHOTOMETER FOR ASTRONOMY 199

Be stars:

With R = 100, all known Be stars can be

spectroscopically studied with IRSP. The above make IRSP an extremely versatile

instrument.

7.2 Extended nebulosities in star forming regions

In recent ti mes, several NIR mapping studies of Galactic star forming regions are being carried out

with many interesting resu lts. In addition to embedded ste llar components, the nebulosities near young ste ll ar objects (YSOs), Herbig Haro objects,

etc . are a hot topic of research . They are often assoc iated wi th strong outflow activity. The IRSP is also capable of undertak ing similar studies. One

example is: The nebular emi ssion extends over - 35" for the L 125 1 molecular cloud with ongoing star formation, wh ich has been studied20

. Typical intensity in K band is 14-15 mag/sq. arc sec. In a 5"><5" pixel , expected K is 11 .5 mag, which is well within the

capability of IRSP.

7.3 Reflection nebula

Recent studies of samples of reflection nebulae have demonstrated the role of non-equilibrium emission from very small grains (VSGs) and polycycl ic aromatic hydrocarbons (PAHs) (Ref. 21). This emission leads to a strong continuum as well as

PAH features in the 1-20 ~m wavelength region22 .

The emi ss ion regions are typically 6-10" in size; integrated K magnitude is - 13 . It has been concluded

that the VSG or PAH continuum emission is much stronger than the scattered star lighe3

. Using the IRSP, several such objects can be studied to verify

the above hypothesis. The UV pumped fluorescence model for such emission features can also be tested.

7.4 Flare monitor

With the advent of Hubble space telescope (along with some terrestrial telescopes), in recent times, many interesting flaring actIVItIes have been discovered . Of particular interest -are the dMe flare stars like AD Leo, AU Mic, etc . Recent discovery of their fl aring in UV lines24

-26 has made them targets of

many studies. Their NIR behaviour during flaring in the UV lines will be extremely useful information

which is lacking at present. Many of these objects are

bright enough to be monitored by IRSP, and they can form an important science objective for IRSP.

Time variability in NIR fer certain other class of

stars also is quite interesting. These are: Mira variables; Asymptotic giant branch (AGB) stars; and RS CVn stars. The monitoring -of NIR colours of

Miras over several cycles will be an important input to the modelling of their atmospheres. Similarly, the study of NIR variability of AGB stars will lead to the understanding of circumstellar shells around them. The RS CVn stars have active spots on their surface which are being studi ed even in the x-ray band. Their

NIR light curve will also constitute an interesting

sc ience programme for IRSP.

7.5 Novae light curve

Typically a few novae occur in our Galaxy every year, which can be easily detected in optical/infrared or even x-rays. The current estimates suggest that the dust formation takes place about 2-3 months after the optical peak. This can be studied and verified from the observed NIR light curves. Comparison of these

measurements with optical data can lead to much better understanding of the material ejected during the novae outburst. Hence, the study of novae light curves will form an interesting science goal for IRSP.

7.6 Stellar continuum

Since the IRSP photometer will concentrate in the non-standard NIR bands (in between the atmospheric

windows), a new database of these new infrared colours for various types of stars will become available. The colour-colour studies in these new

bands might spring some surprises, since it will be available for the first time on a large set of stars.

7.7 Polarization mapping

In the aforesaid description of IRSP no polarizer has been proposed in order to keep IRSP instrument free from any moving part. However, permanent inclusion of a polarizer for a particular photometric band may be considered seriously. With this approach, without reducing the reliability, a completely new dimension can be offered to the possible science with IRSP. Of course, as a price,

200 INDIAN J RADIO & SPACE PHYS, AUGUST 1999

some sensitivity has to be sacrificed. If this channel is

made independent of the other main photometer

channels, the NIR polarimetry may be viable in IRSP.

8 Conclusions A space-borne near infrared spectro-photometer

(IRSP) has been described, whose technical

feasibility is within the reach of the Indian Space

Research Organization (lSRO) and astronomical

research laboratories within the country. Several of

the astrophysical problems that can be meaningfully

addressed by IRSP have been described . This

instrument (IRSP) is expected to be very reliable by

keeping it s imple with no moving parts and can be planned as a piggyback payload on any Indian

mi ssion in near future. In spite of its simplicity, IRSP

can achieve interesting and internationally

competitive sc ience.

Acknowledgements The author thanks Prof. P C Agrawal and Dr K

Kasturirangan (Chairman , ISRO) for enthusing the

entire space astronomy community in the country by

strongly supporting the idea of Indian multi­

wavelength astronomical satellite (IMAS). The

present homework is a result of (he two-day national

level meeting arranged by them at ISRO Head

Quarters, Bangalore, to critically discuss various

possibilities about IMAS . Thanks are also due to

Prof. S N Tandon and Prof. D Narasimha for several

useful di scussion . The anonymous referees are

thanked for their suggestions, which improved this

contribution .

References

Kasturirangan K, Marar T M K, Padmi ni V M, Prasad N L, Rao U R & Scetha S, Astron & Astraphys (Germany), 283 (1994) 435 .

2 Agrawal P C, Paul B, Rao A R, Shah M R, Mukerjee K, Vahi a M N. Yadav 1 S, Dedhia D K, Malkar i P, Shah P B, Daml e S V. Marar T M K, Seetha S, Chitnis V R, UpaLih yaya N. Sharma R K, Murthy N S, Abraham L & Ka"'Jrirangan K, J Korean Astran Soc (South Korea j, 29 ( 1997) S429.

3 Agrawal P C, Hi!(h Energy Astronomy & Astrophysics, edited hy P C Agrawal & P R Vishwanath (U niversities Press, Hydcrabad) , 1998, p408.

4 Marar T M K, Sharma M R, Seetha S, Upadhyaya N, Padmini V N, Umapathy C N, Prasad N L, Murthy N S R,

Sreenivasaiah K V & Kasturirangall K, Astran & Astrophys (Gennany), 283 (1994) 698.

5 Paul B, Agrawal P C, Rao A R, Vahi a M N, Yadav J S, Marar T M K, Seetha S & Kasturirangan K, Astron & Astraphys (Germany), 320 (1997) L37.

6 Rao A R, Agrawal P C, Paul B, Vahia M N, Yadav J S, Marar T M K, Seetha S & Kasturirangan K, Astron & Astrophys (Germany), 330 (1998) 18 1.

7 Schember H, The Next Generation Space Telescope (a NASA & STScI publication, USA), 1989, p285.

8 Schlessinger M, Infrared Technology Fundamentals (Marcel Dekker Inc., New York, USA) , 1995.

9 Thompson R I. Proc S PIE ·· lnt Soc Opt Eng (USA) , 2 198 (1994) 1202.

10 Thompson R I, The Next Generation Space Telescope (a NASA & STScI publication , USA), 1989, p310.

II McLean I S, Infrared Astronomy, edi ted by A Mampaso, M Prieto & F Sanchez (Cambridge University Press, Cambridge), 1993 , p343 .

12 Neugebauer G, Habing H J, van Duinen R, Aumann H H, Baud B, Beichman C A, Beintema D A, Boggess N, Clegg P E, Jong T, Emerson J P, Gautier T N, Gillett F C, Harris S, Hauser M G, Houck J R, Jennings R E, Low F J, Marsden P L, Miley G, Olnon F M, Pottasch S R, Raimond E, Rowan­Robinson M, Soifer B T, Walker R G, Wessel ius P R & Young E, ASlraphys J (USA), 278 (1984) LI.

13 Boggess N W, Mather J C, Weiss R, Bennett C L , Cheng E S, Dwek E, Gulki s S, Hauser M G, Janssen M A, Kelsall T, Meyer S S, Moseley S H, Murdock T L, Shafer R A, Silverberg R F, Smoot G F, Wilkinson D T & Wright E L, Astrophys J (USA), 397 (1992) 420.

14 Murakami H, Freund M M, Ganga K, Guo H, Hirao T, Hiromoto N, Kawada M, Lange A E, Makiuti S, Matsuhara H, Matsumoto T, Matsuura S, Murakami M , Nakagawa T, Narita M, Noda M, Okuda H, Okumura K, Onaka T, Roellig T L, Sato S, Shibai H, Smith B i, Tanabe T, Tanaka M, Watabe T, Yarnamura I & Yuen L, Pub{ ASlron Soc Jpn (Japan ), 48 (1996) L41.

15 Kessler M F, Steinz J A, Anderegg M E, Clavel i , Drechsel G, Estaria P, Faelker J, Riedinger J R, Robson A, Taylor B G & Ferran S X, Astron & Astraphys (Germany), 315 (1996) L27.

16 Thompson R I, Rieke M, Schneider G, Hines D C & Corbin M R, ASlrophys J (USA), 492 ( 1998) L95.

17 Werner M W & Simmons L L, Space Sci Rev (Netherlands) , 74 (1995) 125.

18 Hills R, Space Sci Rev (Netherlands), 74 (i 995) 119.

19 Pati A K & Rao N K, Th e Ultra· Violet Imaging Telescope (A proposal for an astronomy initiat ive in space in the 21st century, India), 1998.

20 Rosvick J M & Davidge T J, Pub AstrolJ Soc Pac (USA), 107 (1995) 49.

21 Sellgren K, Werner M W & Allamandola L J, ASlrop/iys J Supp Ser (USA), 102 (1996) 369.

22 Sellgren K, Luan L & Werner M W, Astraphys J (USA), 359 (1990) 384.

23 Sellgren K, Werner M W & Dinersiein H L, Astrophys J (USA), 400 (1992) 238.

GHOSH; SPACE-BORNE INFRARED SPECTRO-PHOTOMETER FOR ASTRONOMY 201

24 Bookbinder J A, Walter F M & Brown A, Cool stars, stellar systems and the Sun in Astronomical Society of the Pacific Conference Series, Vol. 26,1992,1>27.

25 Robinson R D. Shore S N, Carpenter K G, Woodgate B E, Maran S p, Brandt J C, Kundu M.R, White S M , Linsky J L & Walter F M, Cool stars, stellar systems and the Sun in

Astronomical Society of the Pacific Conference Series, Vol. 26, 1992, P 31.

26 Maran S P, Robinson R D, Shore S N, Brosins J W, Carpenter K G, Woodgate B E, Linsky J L, Brown A, Byrne P B, Kundu M R, White S, Brandt J C, Shine R A & Walter F M, Astrophys J (USA), 421 ( 1994) 800.