Detection of dust in the most distant known radiogalaxy

a r X i v :a s t r o -p h /9506039v 1 6 J u n 1995

Mon.Not.R.Astron.Soc.000,000–000(0000)Printed 1February 2008

(MN L A T E X style ?le v1.4)

Detection of dust in the most distant known radiogalaxy

R.J.Ivison

Royal Observatory,Blackford Hill,Edinburgh EH93HJ

Accepted 1995May ??.Received 1995May ??;in original form 1995March 1

ABSTRACT

A search for millimetric continuum emission from eight optically-selected,radio-quiet

quasars and a radiogalaxy with 3.7

1INTRODUCTION

There are approximately 50known quasars and radiogalax-ies (RGs)with z ～4.It is natural to employ these objects as observational probes of the early stages of galaxy formation;for example,low-redshift RGs represent a relatively homo-geneous group in many respects,and comparison with RGs at higher redshifts gives some perception of their evolution-ary traits,always bearing in mind fears about sample bias,e.g.the alignment e?ect (McCarthy et al.1987;Chambers,Miley &Van Breugel 1987).

Young quasars and RGs are likely to be the sites of ac-tive star formation,and they are therefore an obvious place to search for thermal emission from dust and line emission from molecular gas (McMahon et al.1994).Evidence already supports the idea that there are vast reservoirs of dust-rich gas in several high-redshift objects,notably in the radio-quiet quasar BR1202?0725(Isaak et al.1994;McMahon et al.1994),the Cloverleaf quasar (Barvainis et al.1994),and the RG 4C41.17(Dunlop et al.1994;Chini &Krugel 1994),all of which have been detected in the rest-frame far-infrared.A thermal origin for the millimetric emission from these objects would indicate dust masses of around 108–109M ⊙,and it is hard to explain their submillimetre spec-tral indices (α～3–4,where F ν∝να)by anything other than emission from dust (e.g.Hughes et al.1993).

Determining the evolutionary phase of these objects

(e.g.whether a ‘proto-galactic’label applies)is not straight-forward.One could measure the fraction of their gas which

has been converted into stars,but this relies on knowledge of the relative gas-to-dust ratios over a range of redshifts,which emphasizes the importance of both continuum and molecular-line data.The millimetric continuum data pre-sented here will be complemented by centimetric spectral-line observations of CO (Ivison et al.,in preparation),which should yield estimates of several important physical param-eters,not least the mass and temperature of the gas and dust.

2BOLOMETER ARRAY MEASUREMENTS

The data were obtained during 1995February 17–22us-ing the MPIfR 7-channel 3He-cooled bolometer array (see Kreysa 1993)and the 30-m IRAM Millimetre Radio Tele-scope on Pico Veleta,Spain.The individual beamsizes are 11–12arcsec (HPBW),and the bolometers are separated by 22arcsec in a hexagonal arrangement surrounding a central pixel.During our observing session we employed a 1-arcmin azimuthal chop-throw,at a rate of 2Hz;the telescope was also position-switched by 1arcmin every 10s in the stan-dard symmetric ON-OFF-OFF-ON mode.The nett result was that one third of the total observing time was spent on the source.

c

0000RAS

2R.J.Ivison

Table1.1.25-mm photometry.

Source Object z Adopted coordinates(B1950)λ0UT Date Flux Density Note

name typeαδ/μm(1995Feb)(±σ)/mJy/Reference

PC0027+0521RQQ 4.21002730.10+052139.024020.6?0.27±4.661,SSG94

BRI0151?0025RQQ 4.20015105.90?002544.024020.70.71±2.251,S94

PC1301+4747RQQ 4.00130150.80+474735.025021.21.20±1.77SSG91,SSG94,SVSG92

8C1435+635RG 4.26143527.50+633212.823817.33.03±0.852,3,L94,S95

19.32.61±0.703,4

22.22.26±0.673,5

PC1640+4628RQQ 3.70164037.20+462801.026618.31.08±1.13SSG91,SSG94,SVSG92 PC1643+4631A RQQ 3.79164333.50+463138.026118.50.11±1.006,SSG91,SSG94,SVSG92 PC1643+4631B RQQ 3.83164352.40+463102.025919.5?0.95±0.81SSG91,SVSG92

PC1644+4744RQQ 3.70164448.40+474424.026620.5?0.03±0.95SSG94

PC2047+0123RQQ 3.80204750.70+012356.026017.51.51±0.917,SSG91,SVSG92

19.61.97±1.197

22.42.11±0.645,7

1:high rms due to poor weather;2:τ=0.31;3:weighted mean is2.57±0.42mJy(6.1σ);4:τ=0.12;5:τ=0.08;6:5-arcsec SW of SSG94position;7:weighted mean is2.08±0.47mJy(4.4σ);L94:Lacy et al.(1994);S94:Smith et al.(1994);S95:Spinrad et al.

(1995);SSG91:Schneider,Schmidt&Gunn(1991);SSG94:Schneider,Schmidt&Gunn(1994);SVSG92:Schneider et al.(1992).

Typically15–20samples were obtained for each of the targets(separated by pointing,focusing,opacity tips and calibration scans of Uranus and Mars,for which T b of101 and205K were assumed,respectively),each consisting of forty10-s sub-samples.The weighted mean of the outer channels was subtracted from the central channel—an ef-fective method of reducing the e?ects of sky-noise,i.e.the fraction of atmospheric emission that remains after chop-ping.The samples were then concatonated,tested for spikes, and corrected both for atmospheric attenuation and for a gain-elevation dependence,G el,of the form:

G el=(cos(el?45?)2+0.4(sin(el?45?)2)?1.(1)

The sky transparency at1.25mm was virtually con-stant during each session;it varied from night to night,but was generally good,with the opacity ranging from0.06to 0.44.All the sources were observed at low zenith distances (<45?).The pointing characteristics of the telescope were excellent,with rms?uctuations at20per cent of a single-channel HPBW.

The z=4.26RG,8C1435+635,and the z=3.80 quasar,PC2047+0123,were each observed on three sepa-rate occasions in order to con?rm that their emission was real.The?ux-density measurements are presented in Ta-ble1,together with redshifts,rest wavelengths and adopted coordinates.

3RESULTS AND DISCUSSION

The data from Table1reveal that8C1435+635,a z=4.26 RG(Lacy et al.1994,hereafter L94;Spinrad et al.1995, hereafter S95),was detected at the6.1-σlevel,and the radio-quiet quasar,PC2047+0123at z=3.80,was marginally (4.4σ)detected.Limits of3σ<3mJy have been set for four quasars,together with higher limits for a further three.

3.18C1435+635

The integrated spectrum of8C1435+635steepens fromα=?1.2to?2.2via?1.5between0.151,1.5,4.9and8.4GHz;it then continues atα=?2.2to15GHz.8C1435+635was initially selected for blind spectroscopic observations by L94 on the basis of its steepening radio spectrum.

Fig.1shows the measurement obtained here for 8C1435+635,and those obtained by L94and S95at lower and higher frequencies.Flux densities upper limits from the Infrared Astronomical Satellite have also been estimated at 12,25,60and100μm by searching a1square degree?eld centred on the RG for sources from the Faint Source Cat-alogue(FSC),adopting the faintest in each band as the upper limit.This crude method relies on the fact that if the FSC’s sophisticated search routines cannot?nd a point source,then the source must be below the3-σthreshold. The method is probably as reliable as any other,and is less prone than some to providing misleadingly low limits.

If the cm-wave radio spectrum continues to decline above15.2GHz withα=?2.2then the non-thermal con-tribution at243GHz would be15μJy.Between15and 243GHz,α=?0.3,making it a safe bet that its contri-bution to the243-GHz?ux density is completely insigni?-cant,and that the243-GHz emission is probably dominated by thermal emission from dust.Of course,it is not possible to discriminate against a synchrotron origin,and absolute con?rmation of the emission mechanism must be obtained at frequencies above or below243GHz;the most reliable method would be to obtain photometry at0.8mm using UKT14on the15-m James Clerk Maxwell Telescope,where the thermal emission contribution would be around15mJy forα=+4.0(contributions of+2.0from both the dust emissivity and the Rayleigh-Jeans law)—within the capa-bilities of the instrument on an excellent night(e.g.Dunlop et al.1994).

If the243-GHz emission mechanism is thermal,and as-suming a range of estimates for the dust temperature based on the z=3.8RG,4C41.17,and on samples of low-redshift quasars and RGs(Knapp&Patten1991;Andreani,La Franca&Cristiani1993;Chini&Kr¨u gel1994;Dunlop et al. 1994),i.e.T d=20–100K,and an Einstein-de Sitter uni-verse where H0=50km s?1Mpc?1,then for optically thin emission we can determine the dust mass using(from Chini &Kr¨u gel1994):

c 0000RAS,MNRAS000,000–000

Dust in8C1435+6353 Figure1.Spectral energy distribution of8C1435+635.The line drawn in the rest-frame far-IR illustrates a60-K greybody,with+2 frequency dependence for the dust-grain emissivity(neither parameter is reliably constrained).

M d=Fν

obs

D2L

H0

.(3)

For T d=60K,the resulting dust mass is1.6×108M⊙, which is similar to that calculated for4C41.17using the same method.For20

Dust-rich gas in these quantities may explain the anomalously weak Lymanαluminosity of this RG,as sug-gested by L94.Following Chini&Kr¨u gel(1994),in the ide-alized case where108M⊙of dust is distributed in a50-kpc diameter sphere,the optical depth at1216?A would be1–2, su?cient to entirely absorb the Lymanαphotons assuming that neutral hydrogen is su?ciently abundant to make res-onant scattering important(where scattering increases the path length and,accordingly,the optical depth to absorption by dust;see Eales&Rawlings1993).This does not,how-ever,explain why the Lymanαluminosity of8C1435+635 is low relative to4C41.17,another distant,dust-rich radio-galaxy.S95note that8C1435+635lies within the scatter of the Lymanαluminosity—radio power relationship.

3.2Quasars

We observed a total of eight radio-quiet quasars,all of which satisfy3.7

Upper limits(3σ)of225and120μJy were obtained for PC2047+0123at1.5and4.9GHz by Schmidt et al. (1995)and Schneider et al.(1992),respectively;yielding a lower limit for the spectral index between4.9and243GHz ofα>+0.7.This provides evidence,always bearing in mind the marginal nature of the detection and the lack

c 0000RAS,MNRAS000,000–000

4R.J.Ivison

of a measurement of the submillimetre spectral index,that the1.25-mm?ux density is dominated by thermal emission from dust,though again we cannot discriminate against a non-thermal mechanism.If dust is responsible,we estimate M d=1.5×108M⊙(under the assumptions discussed in the previous section,and withκν

=8.7cm2g?1).

4CONCLUDING REMARKS

In an attempt to increase the number of known gas-rich systems at z～4,eight radio-quiet quasars and a RG,which satisfy3.7

If the1.25-mm emission from8C1435+635is dominated by thermal emission,which seems likely given the steep cen-timetric spectrum,then the derived dust mass is similar to those found in nearby radio-quiet quasars(Hughes et al. 1993),and1–2orders of magnitude higher than those found for nearby radio galaxies(e.g.Knapp,Bies&van Gorkom 1990;Knapp&Patten1991).There is evidence,therefore, that spectral-line and sub-mm continuum observations of this RG will provide clues concerning the dynamical,physi-cal and evolutionary state of some of the most ancient known material in the Universe,and there is a tantalizing possibil-ity that the Very Large Array can be used to map the molec-ular gas with the same angular resolution as can be routinely obtained when observing CO in local galaxies(0.1arcsec～1kpc at z～4,for?0=0.5and H0=50km s?1Mpc?1).

ACKNOWLEDGMENTS

It is a pleasure to acknowledge the excellent support of Raphael Moreno at Pico Veleta.

REFERENCES

Andreani P.,La Franca F.,Cristiani S.,1993,MNRAS,261,L35 Barvainis R.,Tacconi L.,Antonucci R.,Alloin D.,Coleman P., 1994,Nat,371,586

Chambers K.C.,Miley G.K.,van Breugel W.J.M.,1987,Nat,329, 624

Chini R.,Kr¨u gel E.,1994,A&A,288,L33

Dunlop J.S.,Hughes D.H.,Rawlings S.,Eales S.,Ward M.,1994, Nat,370,347

Eales S.A.,Rawlings S.,1993,ApJ,411,67

Hales S.E.G.,Masson C.R.,Warner P.J.,Baldwin J.E.,1990,MN-RAS,246,256

Hughes D.H.,Robson E.I.,Dunlop J.S.,Gear W.K.,1993,MN-RAS,263,607

Isaak K.G.,McMahon R.G.,Ellis R.E.,Withington S.,1994,MN-RAS,269,L28

Knapp G.R.,Bies W.E.,van Gorkom J.H.,1990,AJ,99,476 Knapp G.R.,Patten B.M.,1991,AJ,101,1609Kreysa E.,1993,in Guyenne T.D.,ed.,Photon Detectors for Space Instrumentation.ESA/ESTEC,Noordwijk

Kr¨u gel E.,Steppe H.,Chini R.,1990,A&A,229,17

Lacy M.et al.,1994,MNRAS,271,504(L94)

McCarthy P.J.,van Breugel W.J.M.,Spinrad H.,Djorgovski S., 1987,ApJ,321,L29

McMahon R.G.,Omont A.,Bergeron J.,Kreysa E.,Haslam

C.G.T.,1994,MNRAS,267,L9

Rees N.,1990,MNRAS,244,233

Schmidt M.,van Gorkom J.H.,Schneider D.P.,Gunn J.E.,1995, AJ,109,473

Schneider D.P.,Schmidt M.,Gunn J.E.,1991,AJ,101,2004 Schneider D.P.,Schmidt M.,Gunn J.E.,1994,AJ,107,880 Schneider D.P.,van Gorkom J.H.,Schmidt M.,Gunn J.E.,1992, AJ,103,1451

Smith J.D.et al.,1994,AJ,108,1147

Spinrad H.,Dey A.,Graham J.R.,1995,ApJ,438,L51(S95) This paper has been produced using the Royal Astronomical Society/Blackwell Science L A T E X style?le.

c 0000RAS,MNRAS000,000–000