Hard X-ray Emission From Low Mass X-ray Binaries

超短超强激光辐照靶物质产生K-alpha源王向贤【摘要】超短超强激光与物质相互作用产生的K-alpha线辐射,有准单能、发射区域小、时间短等优点,具有广泛的应用前景.介绍了超短超强激光辐照靶物质产生K-alpha源的基本原理及其主要研究内容,讨论了该领域的研究热点.【期刊名称】《巢湖学院学报》【年(卷),期】2011(013)003【总页数】4页(P45-47,110)【关键词】超短超强激光;K-alpha源;基本原理【作者】王向贤【作者单位】巢湖学院物理与电子科学系,安徽巢湖238000【正文语种】中文【中图分类】O434超短超强激光与物质相互作用产生的K-alpha线辐射。

有准单能（几十个keV）、发射区域小（微米量级），时间短（飞秒-皮秒量级）等优点[1,2]。

可广泛应用于惯性约束聚变背光照相，医学成像，光刻，时间分辨X射线衍射等领域。

同时，超短超强激光与物质相互作用中超热电子辐射是强场物理的重要研究内容之一，而K-alpha线的产生和超热电子直接相关，故可以通过研究K-alpha线辐射研究超短超强激光与物质相关作用产生的超热电子。

如图1所示，用超短超强激光脉冲辐照靶物质，如铝（Al）、铜（Cu）、金（Au）等，激光与靶物质的耦合将产生超热电子，超热电子向靶中输运，碰撞电离1S轨道电子，使得1S轨道产生空穴，此时2P轨道电子将向1S轨道跃迁，产生K-alpha光子辐射，产生的K-alpha线辐射包括K-alpha1线和K-alpha2线[3]，分别对应于跃迁22P3/2→12S1/2 和22P1/2→12S1/2。

基于超短超强激光脉冲驱动的K-alpha源的实验布局如图2所示。

主激光经全反射镜反射后，被离轴抛面镜聚焦到铜等靶物质上。

X射线光谱仪（如：光子计数型CCD、晶体谱仪等）用于测量K-alpha线光谱，安装在与入射激光处于同一水平面的靶室法兰上（靶前、靶后位置均可），电子谱仪可同时在线测量实验产生的超热电子能谱。

a r X i v :a s t r o -p h /9707086v 2 21 J u l 1997Hard X-ray Emission from Cassiopeia A SNRLih-Sin The ∗,Mark D.Leising ∗,Dieter H.Hartmann ∗,James D.Kurfess †,Philip Blanco §,and Dipen Bhattacharya ‡∗Department of Physics and Astronomy,Clemson University,Clemson,SC 29634-1911†E.O.Hulburt Center for Space Research,Naval Research Laboratory,Code 7650,Washington,DC 20375-5352§Center for Astrophysics and Space Sciences,University of California,San Diego,CA 92093‡Institute of Geophysics and Planetary Sciences,University of California,Riverside,CA 92521Abstract.We report the results of extracting the hard X-ray continuum spectrum of Cas A SNR from RXTE/PCA Target of Opportunity observations (TOO)and CGRO/OSSE observations.The data can rule out the single thermal bremsstrahlung model for Cas A continuum between 2and 150keV.The single power law model gives a mediocre fit (∼5%)to the data with a power-law index,Γ=2.94±0.02.A model with two component (bremsstrahlung +bremsstrahlung or bremsstrahlung +power law)gives a good fit.The power law index is quite constrained suggesting that this continuum might not be the X-ray thermal bremmstrahlung from accelerated MeV electrons at shock fronts [1]which would have Γ≃2.26.With several SNRs detected by ASCA showing a hard power-law nonthermal X-ray continuum,we expect a similar situation for Cas A SNR which has Γ=2.98±0.09.We discuss the implication of the hardest nonthermal X-rays detected from Cas A to the synchrotron radiation model.INTRODUCTIONX-ray observations of supernova remnants have been stimulated by recent re-ports of nonthermal power-law X-ray detections suggesting supernovae as sites of charged particle accelerations and sources of cosmic rays.The first strong evidence for charged particle acceleration near supernova shock fronts in the X-ray energy band is demonstrated by the morphological and spectral correlation between X-ray and radio emission from the bright NE and SW rims of SN 1006[2].The brightest radio emission regions show almost featureless power law X-ray spectra when com-pared with SN 1006central region which is dominated by emission lines of highly ionized elements in a non-equilibrium ionization thermal plasma.The nonthermal X-ray component has been modeled as a synchrotron emission from electrons ac-FIGURE1.The thermal bremsstrahlung(kT=7.93keV;dashed line)+power-law(Γ=2.98; dashed-dotted line)+6K X-ray line model produces a goodfit(solid line)to the PCA&OSSE data(crosses)with aχ2/dof=1.077,dof=62.celerated to∼100TeV within shock fronts[3–5].In this model,the radio spectrum is produced by synchrotron radiation of electrons accelerated to∼GeV energies with the radio power-law index being less steep than the X-ray power-law index. Several similar evidences also have been demonstrated by ASCA measurement of RX J1713.7-3946[6]and IC443[7],and RXTE and OSSE measurements of Cas A [8–10].OSSE with a total accumulation time of15×105detector-seconds,detected a hard continuum between40-150keV from Cas A SNR at a4σconfidence with a flux of∼9×10−7γcm−2s−1keV−1at100keV[10].The detection is the hardest X-ray detection from a SNR without plerionic source.However,the shape of the continuum has not been strongly constrained.The continuum can either be a bremsstrahlung with kT≃35keV or a power law withΓ≃3.06.In this paper,we use the2-30keV RXTE/PCA TOO and the40-150keV CGRO/OSSE Cas A data to better determine the shape of Cas A hard X-ray continuum.RESULTSThe RXTE/PCA data used in this analysis is the TOO by RXTE on Aug.2, 1996with a total observing time of∼4000sec.Wefit the2-30keV PCA data simultaneously with the40-150keV OSSE data.The results offitting several models with one or two continuum plus about six emission lines from Si,S,Ar,Ca, and Fe are shown in Table1.The line widths arefixed to100eV and the hydrogen column density toward Cas A isfixed at1×1022H cm−2[11,12].Wefind that a single thermal bremsstrahlung model for the data can be ruled out.Nevertheless, a single power law model gives a mediocrefit.We alsofind the two models that bestfit the EXOSAT[12]and Tenma[11]data cannot give acceptablefits to the PCA and OSSE data.A two component model,either the bremsstrahlung+ bremsstrahlung or the bremsstrahlung+power law,each gives an equally good fit.In Figure1for the current interest in a nonthermal power law model,we show the result of the thermal bremsstrahlung+power-law modelfit to the data.The power law continuum is detected with a higher confidence level and the power-law index is more tightly constrained than using the OSSE data alone[10].The PCA and OSSE data is consistent only at a3.8%level with the X-ray bremsstrahlung from accelerated MeV electrons at shock fronts as suggested by Asvarov et al.[1] which would haveΓ=α+1.5=2.26whereαis the radio spectral index.FIGURE2.The synchrotron photon energy relation with the electron energy[2]for Cas A(solid line),IC443(dashed line),and SN1006(dashed-dotted line).The solid squares and the open squares mark the maximum electron energy constrained by the synchrotron energy loss(Eq.(1) of[4])for f*R J=1and f*R J=10,respectively.The solid triangles and the open triangles mark the maximum electron energy bounded by the age of the SNRs(Eq.(2)of[4])for f*R J=1and f*R J=10,respectively.This maximum energy is shown only for SN1006due to its low magnetic field.Synchrotron energy loss sets the maximum electron energy in Cas A and IC443.The shock speed is assumed to be5000km/s in each case.f is the ratio of the electron mean free path to the electron gyroradius.R J is the factor that contains the orientation of the shock relative to the magneticfield.The dotted lines shows some interesting synchrotron photon energies at1GHz, 10keV,and100keV.A summary of the detected nonthermal continuum from four SNRs is shown in Table2.Cas A is of interest because from the4SNRs with measured nonther-mal power-laws,it is the youngest,has the strongest magneticfield,its detected nonthermal energy is the highest,and it has the steepest X-ray and radio power-law indexes,of four SNRs with measured nonthermal power law.In Figure2,we show for Cas A,IC443,and SN1006how close their detected hard X-ray energies to the maximum synchrotron energies.The synchrotron energy is related to the maximum electron energy which is limited by the synchrotron energy loss and the SNR’s age.Cas A,as the brightest radio SNR source,has the most intense mag-neticfield and therefore it causes the large synchrotron energy loss which limits Cas A’s maximum electron energy.Figure2shows that the hard X-ray emission detected from Cas A by OSSE is near the cutoffsynchrotron energy or indirectly to the maximum electron energy.Further Cas A measurements by OSSE may detectparison of the measured nonthermal continuumSN10069713-10µG0.560.5-20 2.95RX J1713.7-3946≥1000??0.5-10 2.45IC4431000500µG<0.240.5-12 2.35Cas A3171-5mG0.7615-150 2.98the cutoffenergy and hence would provide the evidence that the nonthermal power law emission is the synchrotron radiation from accelerated electrons at the shock fronts near the maximum energy.Another evidence of the synchrotron radiation being the process for the nonthermal emission is the steepening of spectral index from radio to X-ray energy.It has been suggested that the turnover or break∼0.25 keV is observed in SN1006spectra[5].We expect the turnover in Cas A spectra is∼1-10keV due to Cas A’s strong magneticfield.However,in order to observe this turnover in Cas A spectra,an X-ray imaging detector is needed because Cas A’s thermal bremsstrahlung contributes substantially near this energy.REFERENCES1.Asvarov,A.I.,Guseinov,O.Kh.,Dogel,V.A.,Kasumov,F.K.,Soviet Astron.,33,532(1989)2.Koyama,K.,Petre,R.,Gotthelf,E.V.,Hwang,U.,Matsuura,M.,Ozaki,M.,andHolt,S.S.,Nature,378(1995)3.Ammosov,A.E.,Ksenofontov,L.T.,Nikolaev,V.S.,and Petukhov,S.I.,Astron.Lett.20,157(1994)4.Reynolds,S.P.,ApJ,459,L13(1996)5.Mastichiadis,A.,A&A,305L53(1996)6.Koyama,K.,Kinugasa,K.,Matsuzaki,K.,Nishiuchi,M.,Sugizaki,M.,Torii,K.,Yahauchi,S.,and Aschenbach,B.PASJ,49,in press(1997)7.Keohane,J.W.,Petre,R.,Gotthelf,E.V.,Ozaki,M.,and Koyama,K.,ApJ,in press(1997)8.Rothschild,R.,et al.,this proceeding(1997)9.Allen,G.E.,et al.,submitted to ApJL(1997)10.The,L.-S.,Leising,M.D.,Kurfess,J.D.,Johnson,W.N.,Hartmann,D.H.,Gehrels,N.,Grove,J.E.,and Purcell,W.R.,Astron.Astrophys.Suppl.120,357(1996) 11.Tsunemi,H.,Yamashita,K.,Masai,K.,Hayakawa,S.,and Koyama,K.,ApJ,306,248(1986)12.Jansen,F.,Smith,A.,Bleeker,J.A.M.,De Korte,P.A.J.,Peacock,A.,and White,M.E.,ApJ,331,949(1988)13.Woltjer,L.,Ann.Rev.Astr.Ap.,10,129(1972)14.McKee,C.,ApJ,188,335(1974)15.Borkowski,K.J.,Szymkowiak,A.E.,Blondin,J.M.,and Sarazin,C.L.,ApJ,466,866(1996)。

Ionizing radiationFrom Wikipedia, the free encyclopediaRadiation hazard symbol.Ionizing (or ionising) radiation is radiation with sufficient energy to remove an electron from an atom or molecule. This ionization produces free radicals, atoms or molecules containing unpaired electrons, which tend to be especially chemically reactive.电离辐射是一种能够导致原子或分子中的电子移动的一种辐射。

这种电离将形成游离基以及包含有未配对电子的原子和分子，会产生特殊的化学反应。

The degree and nature of such ionization depends on the energy of the individual particles (including photons), not on their number (intensity). In the absence of heating a bulk substance up to ionization temperature, or multiple absorption of photons (a rare process), an intense flood of particles or particle-waves will not cause ionization if each particle or particle-wave does not carry enough individual energy to be ionizing (an example is ahigh-powered radio beam, which will not ionize if it does not cause high temperatures). Conversely, even very low-intensity radiation will ionize at low temperatures and powers, if the individual particles carry enough energy (e.g., a low-power X-ray beam). In general, particles or photons with energies above a few electron volts (eV) are ionizing, no matter what their intensity.电离的程度和类型由独立粒子（包括光子）的能量决定而并非是它们的数量（密度）。

x射线荧光光谱法英文X-Ray Fluorescence Spectrometry (XRF)。

X-ray fluorescence spectrometry (XRF) is an analytical technique used to determine the elemental composition of materials by measuring the X-rays emitted by the material when it is exposed to a high-energy X-ray beam. This method is widely used in various fields, including geology, environmental science, forensic science, archaeology, and materials science.Principle of Operation.XRF is based on the principle that when a material is irradiated with high-energy X-rays, electrons in the atoms of the material are excited and ejected from their orbits. The resulting vacancies are filled by electrons from higher energy levels, releasing X-rays with energiescharacteristic of the elements present in the material.The energy of the emitted X-rays is specific to each element, and the intensity of the X-rays is proportional to the concentration of the element in the material. By measuring the energies and intensities of the emitted X-rays, it is possible to identify and quantify the elements present in the sample.Instrumentation.A typical XRF spectrometer consists of the following components:X-ray source: Generates high-energy X-rays that bombard the sample.Sample chamber: Holds the sample to be analyzed.Detector: Converts X-rays into electrical signals.Multichannel analyzer (MCA): Digitizes and analyzes the electrical signals from the detector.Types of XRF Spectrometers.There are several types of XRF spectrometers, each with its own advantages and limitations:Energy-dispersive XRF (EDXRF): Uses a solid-state detector to measure the energies of the emitted X-rays. EDXRF is relatively inexpensive and easy to operate, but it has lower energy resolution compared to other types of XRF spectrometers.Wavelength-dispersive XRF (WDXRF): Uses a crystal monochromator to separate the emitted X-rays by wavelength. WDXRF offers higher energy resolution than EDXRF, but it is more complex, expensive, and time-consuming to operate.Total reflection XRF (TXRF): Utilizes total reflection conditions to enhance the sensitivity for analyzing trace elements in liquids. TXRF is highly sensitive, but it requires sample preparation and is not suitable for solid samples.Applications of XRF.XRF is a versatile analytical technique with a wide range of applications:Geochemistry: Determining the elemental composition of rocks, minerals, and soils.Environmental science: Monitoring pollutants in air, water, and soil.Forensic science: Analyzing trace evidence, such as gunshot residue and paint chips.Archaeology: Studying the composition of artifacts and ancient materials.Materials science: Characterizing the elemental composition of metals, alloys, and other materials.Advantages of XRF.Nondestructive: Does not damage the sample being analyzed.Multi-elemental: Can identify and quantify multiple elements simultaneously.Rapid: Provides real-time analysis results.Sensitive: Can detect elements at trace levels.Versatile: Can be applied to various sample types, including solids, liquids, and powders.Limitations of XRF.Limited sensitivity: Cannot detect elements present in very low concentrations.Matrix effects: The presence of other elements in the sample can affect the accuracy of the analysis.Sample preparation: May require sample preparation,such as grinding or homogenization.Cost: XRF spectrometers can be expensive, especially WDXRF systems.Conclusion.X-Ray Fluorescence Spectrometry is a powerful analytical technique that provides valuable information about the elemental composition of materials. It is widely used in various fields and offers advantages such as non-destructiveness, multi-elemental analysis, and rapid results. However, it has limitations in sensitivity and potential matrix effects, which should be considered when selecting this technique for specific applications.。

a rXiv:as tr o-ph/32579v127Fe b23Diffuse X-Ray Emission from the Quiescent Superbubble M 17,the Omega Nebula Bryan C.Dunne 1,You-Hua Chu 1,C.-H.Rosie Chen 1,Justin D.Lowry 1,Leisa Townsley 2,Robert A.Gruendl 1,Mart´ın A.Guerrero 1,and Margarita Rosado 3carolan@ ABSTRACT The emission nebula M 17contains a young ∼1Myr-old open cluster;the winds from the OB stars of this cluster have blown a superbubble around the cluster.ROSAT observations of M 17detected diffuse X-ray emission peaking at the cluster and filling the superbubble interior.The young age of the cluster suggests that no supernovae have yet occurred in M 17;therefore,it provides a rare opportunity to study hot gas energized solely by shocked stellar winds in a quiescent superbubble.We have analyzed the diffuse X-ray emission from M 17,and compared the observed X-ray luminosity of ∼2.5×1033ergs s −1and the hot gas temperature of ∼8.5×106K and mass of ∼1M ⊙to model predictions.We find that bubble models with heat conduction overpredict the X-ray luminosity by two orders of magnitude;the strong magnetic fields in M 17,as measured from H I Zeeman observations,have most likely inhibited heat conduction and associated mass evaporation.Bubble models without heat conduction can explain the X-ray properties of M 17,but only if cold nebular gas can be dynamically mixed into the hot bubble interior and the stellar winds are clumpy with mass-loss rates reduced by a factor of ≥3.Future models of the M 17superbubble must takeinto account the large-scale density gradient,small-scale clumpiness,and strong magnetic field in the ambient interstellar medium.Subject headings:ISM:bubbles —HII regions —ISM:individual (M17)—stars:early-type —stars:winds,outflows1.IntroductionMassive stars dynamically interact with the ambient interstellar medium(ISM)via their fast stellar winds and supernova ejecta.OB associations,with their large concentrations of massive stars,provide an excellent laboratory to study these interactions.The combined actions of the stellar winds and the supernovae from the massive stars in OB associations sweep up the ambient ISM to form expanding shells called superbubbles(Bruhweiler et al. 1980).The physical structure of a superbubble is very similar to that of a bubble blown by the stellar wind of an isolated massive star,as modeled by Castor,McCray,&Weaver (1975)and Weaver et al.(1977).Theoretically,an interstellar bubble consists of a shell of swept-up ISM with its interior filled by shocked fast wind at temperatures of106–108K.There are two basic types of models for wind-blown bubbles:energy-conserving and momentum-conserving.In the former,the shocked stellar wind is separated from the swept-up interstellar shell by a contact discontinu-ity,where heat conduction and mass evaporation may take place.The expansion of the shell is driven by the pressure of the hot interior gas(Dyson&de Vries1972;Castor et al.1975). In the momentum conserving bubbles,the fast stellar winds impinge on the swept-up shell directly,and the expansion is driven by the momentum of the fast stellar wind(Avedisova 1972;Steigman,Strittmatter,&Williams1975).One significant difference between a superbubble and a single star bubble is the possibil-ity that supernovae may occur inside a superbubble and introduce significant perturbations in the surface brightness and characteristic temperature of the X-ray emission,especially if a supernova explodes near the dense shell(Mac Low&McCray1988).This intermittent X-ray brightening has been observed in superbubbles in the Large Magellanic Cloud(LMC). Using Einstein and ROSAT observations,Chu&Mac Low(1990)and Dunne,Points,& Chu(2001)have reported diffuse X-ray emission from a large number of superbubbles in the LMC,and their X-ray luminosities all exceed the luminosities expected by Weaver et al.’s bubble model,indicating recent heating by supernovae.No LMC superbubbles in a quies-cent state,i.e.,without recent supernova heating,have been detected in X-rays by ROSAT (Chu et al.1995).It would be of great interest to detect diffuse X-ray emission from a qui-escent superbubble and compare it to model expectations,as this could provide a valuable diagnostic of bubble models.The emission nebula M17(from the catalog of Messier1850,α=18h21m,δ=−16◦10′(J2000.0),also known as the Omega Nebula,the Horseshoe Nebula,and NGC6618)is located on the eastern edge of a massive molecular cloud,M17SW(Lada,Dickinson,&Penfield 1974).M17exhibits a“blister-like”structure with an overall diameter of∼20′–25′,or∼10–12pc at an adopted distance of1.6kpc(Nielbock et al.2001).The nebula encompassesan open cluster with a stellar age of∼1Myr(Hanson,Howarth,&Conti1997).The open cluster is located on the western side of the nebula,which borders the molecular cloud. Arcuatefilaments extend eastward,creating a shell morphology,and suggesting that it is a young superbubble blown by the OB stars within.The young age of the cluster in M17 implies that no supernova explosions have occurred,thus M17provides an ideal setting to study the generation of hot gas solely by fast stellar winds inside a superbubble at a quiescent state.Recent Chandra observations of M17have revealed diffuse X-ray emission in the vicinity of the embedded open cluster(Townsley et al.2003).As a matter of fact,diffuse X-ray emis-sion from M17over a more extended area was previously detected in a ROSAT observation but was never reported.We have analyzed this diffuse X-ray emission from the interior of M17detected by ROSAT to determine the physical properties of the hot interior gas,and considered bubble models with and without heat conduction.Wefind that models with heat conduction produce results with the largest discrepancy from the observed X-ray luminos-ity and hot gas temperature and mass of the superbubble in M17.This paper reports our analysis of the ROSAT observations of M17and comparisons to a range of bubble models.2.Observations and Data Reduction2.1.ROSAT Archival DataWe have used an archival ROSAT Position Sensitive Proportional Counter(PSPC) observation to study the diffuse X-ray emission from M17and investigate the physical prop-erties of the hot,shocked gas interior to the superbubble.The PSPC is sensitive to X-rays in the energy range0.1–2.4keV and has an energy resolution of∼40%at1keV,with a field of view of∼2◦.Further information on the PSPC can be found in the ROSAT Mission Description(1991).The PSPC observation of M17(sequence number RP500311,PI:Aschenbach)was ob-tained on1993September12–13.It is centered on the nebula atα(J2000)=18h21m04.s78 andδ(J2000)=−16◦10′12.′′0,and has an exposure time of6.7ks.As M17has an angular size of∼20′–25′,the diffuse X-ray emission from the superbubble interior is well contained within the inner window support ring of the PSPC.The PSPC data were reduced using standard routines in the PROS4package under the IRAF5environment.2.2.Optical ImagingTo compare the spatial distribution of the X-ray-emitting gas with that of the cooler ionized gas in M17,we have obtained narrow-band Hαimages of this emission nebula.M17 was observed with the Mount Laguna1-m telescope on2002October28–31using a Tektronik 2K CCD.Afilter with peak transmission at6563˚A and a FWHM of20˚A was used to isolate the Hαline.Because the2K CCD has∼0.′′4pixels and afield-of-view of13.′6,the entire nebula could not be observed in a single exposure.Instead a total of twenty-four300s exposures were acquired at8positions to span the nebula.These exposures were combined to form a mosaic image of the nebula and to reject cosmic-ray events in the individual images using the methods outlined in Regan&Gruendl(1995).3.Analysis of the X-Ray EmissionSignificant X-ray emission is detected from M17in the PSPC observation,as can be seen in Figure1.To determine the nature and origin of this emission,we have analyzed its spatial distribution and spectral properties.We have examined the distribution of the X-ray emission and compared it to the optical morphology of the emission nebula.We have also extracted spectra from the PSPC data and modeled them to determine the physical conditions of the hot gas.3.1.Spatial Distribution of the X-ray EmissionTo study the spatial distribution of the X-ray emission from M17,the data were binned to5′′pixels and then smoothed with a Gaussian function ofσ=4pixels(see Figure1a). We have also taken our Hαmosaic and overlaid it with X-ray emission contours to study the extent of X-ray emission within the H II region(see Figure1b).Diffuse X-ray emission is observed to be well confined by the optical nebula.This diffuse emission shows no evidence of limb brightening,suggesting that the interior of M17is centrallyfilled with hot gas. The peak of the X-ray emission is coincident with the center of the open cluster in M17, where Chandra observations show a large number of point sources superposed on the diffuse emission(Townsley et al.2003).This spatial distribution suggests that the X-ray-emitting hot gas originates in the open cluster,as would be expected in a bubble blown by stellarwinds.As there is a massive molecular cloud to the west of M17,we expect the hot gas to expand more rapidly to the east;thisflow of hot gas to the east has then blown the blister-like bubble seen in optical images.Four additional point sources are detected in thefield around M17.The brightest of these point sources lies to the south of the X-ray peak and has been previously designated 1WGA J1820.6−1615in the WGA Catalog of ROSAT Point Sources(White,Giommi,& Angelini1994).This point source is coincident with OI352(Ogura&Ishida1976),an O8star on the southern edge of the open cluster in M17.The other three point sources lie on the northern edge of the emission nebula.These X-ray point sources,designated as1WGA J1820.8−1603,1WGA J1821.0−1600,and1WGA J1820.8−1556,are coincident with stars GSC606265−01977,SAO161369(a known O5star),and GSC06265−01808, respectively.These point sources are marked in Figure1a.3.2.X-Ray SpectraIn order to examine the spectra of the diffuse X-ray emission,wefirst excluded the point sources found in§3.1.Then,noting that M17is on the edge of a dense molecular cloud,we have sectioned the nebula into four regions to account for anticipated changes in the foreground absorption column density.These regions have been labeled A,B,C,and D and are displayed in Figure2.Additionally,we have selected a background annulus around the superbubble,as indicated in Figure2.The background-subtracted spectra were then extracted from the PSPC eventfiles.The observed X-ray spectra of the superbubble is a convolution of several factors:the intrinsic X-ray spectrum of the superbubble,the intervening interstellar absorption,and the PSPC response function.Because the interstellar absorption and the PSPC response function are dependent on photon energy,we must assume models of the intrinsic X-ray spectrum and the interstellar absorption to make the problem tractable.As the X-ray emission from the superbubble interior appears largely diffuse,we have used the Raymond &Smith(1977)thermal plasma emission model to describe the intrinsic X-ray spectra of the superbubble and the Morrison&McCammon(1983)effective absorption cross-section per hydrogen atom for the foreground absorption,assuming solar abundances for both the emitting and absorbing materials.We then simulated the observed spectrum,combining theassumed models for the intrinsic spectrum and the interstellar absorption with the responsefunction of the PSPC.The best-fit spectrum is found by varying parameters and comparingχ2for the simulated and observed spectra.We performed aχ2grid search of simulated spectralfits to determine the best-fit levelsfor the thermal plasma temperature,kT,and absorption column density,N H.Plots of thebestfits to the X-ray spectra are shown in Figure3,andχ2plots are presented in Figure4.Theχ2plots of regions A,B,and C indicate that the X-ray emission can befit by eitherhigher temperature plasma,∼0.7keV,with lower absorption column density,∼1020−21cm−2,or lower temperature plasma,∼0.2keV,with higher absorption column density,∼1022cm−2.This is a common problem for PSPC spectra with a limited number of counts because ofthe poor spectral resolution and soft energy coverage of the PSPC.The best modelfits favorthe higher temperature plasma and lower absorption column density solution.Indeed,thedetection of soft X-ray emission below0.5keV indicates that the solution with N H∼1022cm−2cannot be valid,as such a solution predicts no significant soft X-ray emission should bedetected.Furthermore,the high absorption column density solution predicts a foregroundabsorption column density for M17equal to the total Galactic H I column density along theline of sight(Dickey&Lockman1990).As M17is located in the plane of the Galaxy atl=15◦03′,b=−00◦40′and has a distance of1.6kpc,we do not expect the majority of theGalactic H I toward this direction to be located in front of the superbubble.From the modelfits,we calculated the unabsorbed X-rayflux,and therefore the X-rayluminosity,L X,of the diffuse X-ray emission from each source region.The normalizationfactor,A,of the thermal plasma model is equal to n e n p dV/4πD2,where n e and n p are the electron and proton number densities,V is the volume of the superbubble,and D is thedistance to the source.Assuming a He:H number ratio of1:10and that the X-ray emittinggas is completely ionized,wefind n e≃1.2n p,and that the volume emission measure can beexpressed as<n2e>fV,where f is the volumefilling factor.We have used the diameter ofthe superbubble,∼10–12pc,as the depth of the X-ray emitting gas in each source region.We determined the volume,V,of each source region by multiplying the surface area of theregion by the depth and a geometric correction factor of2/3(approximated by the volumeratio of a sphere to a cylinder).Taking the volumefilling factor to be f=0.5,we thencalculated rms n e in each region.The best-fit values of kT,A,N H,L X and rms n e are givenin Table1.Note that the modelfit to region D did not converge because it contains a largenumber of unresolved stellar sources as well as diffuse emission;only approximate values forthe X-ray luminosity and absorption column density are given.See Townsley et al.(2003)for a detailed analysis of the Chandra observations of region D.Combining our results from each of the source regions,wefind a total diffuse X-ray lu-minosity of∼2.5×1033ergs s−1in the ROSAT PSPC0.1–2.4keV band,a mean characteristic temperature of kT∼0.72keV or T∼8.5×106K,a mean electron density of∼0.09cm−3,and a total hot gas mass of∼1M⊙.We have calculated the total thermal energy in hot,shocked wind component of the superbubble to be E th∼1×1048ergs with a cooling timescale of t T∼40Myr.Given the stellar age of the open cluster,∼1Myr,we do not expect significant radiative cooling to have occurred.As a rough check,we multiply the current X-ray lumi-nosity by the age of the cluster andfind that the total energy radiated away by the X-ray emission is 10%of the total thermal energy.4.Discussionparisons with Model ExpectationsWe now compare the observed physical properties of M17determined above to theo-retical calculations from basic wind-blown bubble models.Although the M17superbubble is in an inhomogeneous ambient medium with a significant density gradient and the cluster is off-centered(a more complex scenario than is considered in basic bubble models),if the superbubble structure is indeed governed by the physical processes prescribed by these mod-els,we expect the properties of the diffuse X-ray emission to agree with predictions within similar orders of magnitude.Wefirst consider the wind-blown bubble model of Weaver et al.(1977)and will later consider a wind-blown bubble model without heat conduction.4.1.1.A Bubble with Heat ConductionIn the Weaver et al.model,heat conduction and mass evaporation act across the bound-ary between the hot interior gas and the nebular shell to lower the temperature and raise the density of the bubble interior.The temperature and electron density profiles of such a bubble have been calculated by Weaver et al.(1977),and the X-ray luminosities of such bubbles can be determined using two methods outlined by Chu et al.(1995).In thefirst method,we derive the expected X-ray luminosity from the observed physical properties of the gas in the104K ionized shell of swept-up ISM.In the second method,we use the spectral types of massive stars in M17to estimate the combined mechanical luminosity of the stellar winds and then derive the expected X-ray luminosity.These two methods use independent input parameters and thus allow us to check the consistency of the pressure-driven bubble model in addition to comparing the expected and observed X-ray luminosities.X-Ray Luminosity Method1:The Ionized ShellThe expected X-ray luminosity in the ROSAT band of0.1–2.4keV for the Weaver et al. (1977)wind-blown bubble model has been given by Chu et al.(1995),L X=(8.2×1027ergs s−1)ξI(τ)n10/70R17/7pcV16/7km/s,(1)whereξis the metallicity relative to the solar value and in this case we assume a value of unity,I(τ)is a dimensionless integral of value∼2,n0is the number density of the ambient medium in cm−3,R pc is the radius of the superbubble in pc,V km/s is the expansion velocity of the superbubble in units of km s−1.The ambient density n0cannot be measured directly, but assuming that the ram pressure of the expanding shell is equal to the thermal pressure of the ionized superbubble shell,the relation between the ambient density and the density of the ionized shell is given byn0=(9/7)n i kT i/(µa V2exp),(2) where n i is the electron number density in the ionized shell,T i∼104K is the electron temperature in the ionized shell,V exp is the expansion velocity of the bubble,andµa= (14/11)m H(Weaver et al.1977;Chu&Mac Low1990).Adopting a mean electron density of n i∼300cm−3(Felli,Churchwell,&Massi1984)and an observed V exp∼25km s−1(Clayton et al.1985),we calculated an ambient density of n0∼40cm−3.Given the superbubble radius of5–6pc,we have determined an expected X-ray luminosity of∼3×1035ergs s−1.X-Ray Luminosity Method2:Wind Luminosity from OB StarsWe can also calculate the expected X-ray luminosity in an energy-conserving,wind-blown bubble by the following equation from Chu et al.(1995),L X=(1.1×1035ergs s−1)ξI(τ)L33/3537n17/35t19/35Myr,(3)where L37is the mechanical luminosity of the stellar winds in units of1037ergs s−1,and t Myr is the age of the bubble in Myr.To remain independent of Method1,we do not use the value of n0determined for that method.Rather,we use the following relations between ambient density,radius,wind luminosity,bubble age,and expansion velocity,n0=(1.3×108cm−3)L37t3Myr R−5pc,(4)t Myr=(0.59Myr)R pc/V km/s,(5) (Weaver et al.1977;Chu et al.1995).We again take the expansion velocity to be∼25km s−1 (Clayton et al.1985)and the radius to be5–6pc and derive a bubble age of∼0.13Myr.To determine the wind luminosity of M17,we examined its massive stellar content. Hanson et al.(1997)identified nine O stars and four late-O/early-B stars in the open cluster. Using the spectral types of these massive stars,we have estimated their terminal stellar wind velocities,effective temperatures,and luminosities based on the stellar parameters given by Prinja,Barlow,&Howarth(1990)and Vacca,Garmany,&Shull(1996).We then calculated the mass-loss rates for the OB stars in M17by utilizing the empirically derived relationship between effective temperature,luminosity,and mass-loss rate of de Jager,Nieuwenhuijzen,& van der Hucht(1988).Table2lists the detected OB stars,their spectral types as determined from optical and K-band observations,their terminal wind velocities V∞,stellar effective temperatures T eff,stellar luminosities L,and their mass-loss rates˙M.We calculated the total mechanical luminosity of the stellar winds,L w=Σ(1/2)˙MV2∞,(6) from the OB stars to be∼1×1037ergs s−1.As noted by Felli et al.(1984),the identified OB stars can approximately account for the ionization of the emission nebula;we therefore expect our calculated wind mechanical luminosity to be reasonably complete as well.Assuming a relatively constant mechanical luminosity,wefind a total energy deposited by stellar winds of∼4×1049ergs over the life of the bubble.In addition,this value of the wind mechanical luminosity gives an ambient density of∼60cm−3and an expected X-ray luminosity of ∼5×1035ergs s−1.This X-ray luminosity value is consistent to within a factor of two with the value found by Method1.Although the two methods of determining the expected X-ray luminosity are consistent with each other,they do not agree with the X-ray luminosity derived from the PSPC obser-vation.The observed X-ray luminosity is∼100–200times lower than expected from Weaver et al.’s bubble model.It is possible that stellar winds are clumpy,as suggested by Moffat& Robert(1994),then the conventionally derived mass loss rates would be reduced by a factor of≥3.Even using the reduced mass loss rates,the expected X-ray luminosity is more than 40times too high.The observed temperature,density,and surface brightness of the hot gas in M17do not agree with the model expectations,either.The physical conditions of the shocked stellar winds in Weaver et al.’s model are heavily modified by heat conduction and the hot gas mass is dominated by the nebular mass evaporated across the interface.The predicted temperature is5.0–5.6×106K near the center and decreases outward,the predicted density is0.2–0.4cm−3near the center and increases outward,and the X-ray surface bright-ness is expected to show pared with observed properties,the expected temperature is too low,density is too high,and the X-ray morphology is inconsistent.The disagreements between observations and model expectations suggest that heat conduction may not play a dominant role in determining the physical conditions inside this superbubble.Heat conduction can be suppressed by the presence of magneticfields(Soker1994;Band &Liang1988).The magneticfield strength in M17has been measured via the H I Zeeman effect to be100–550µG,peaking near the interface between the H II region and the molecular cloud M17SW(Brogan et al.1999).Assuming a comparable magneticfield strength in the swept-up104K shell,wefind the Alfv´e n speed to be10–60km s−1which is comparable to or much greater than the isothermal sound velocity of the104K gas,10km s−1for H atoms. In addition,the magneticfield strength and isothermal sound velocity indicate a gyro-radius of 10km for protons in the swept-up shell.This suggests that the protons in the104K gas will be unable to escape the magneticfield and diffuse into the interior of the superbubble, inhibiting heat conduction and mass evaporation between the hot interior and the cool shell of the bubble.4.1.2.A Bubble without Heat ConductionWe now turn our consideration to a wind-blown bubble without heat conduction.The X-ray emission of a bubble interior depends on both the temperature and the amount of hot gas.We willfirst compare the plasma temperature expected from the shocked stellar winds to the observed hot gas ing the combined stellar winds mass-loss rate of 4.3×10−6M⊙yr−1(summed from Table2)and the integrated wind mechanical luminosity L w of1×1037ergs s−1as calculated in§4.1.1,we derive an rms terminal wind velocity of V∞∼2700km s−1.The post-shock temperature of the combined stellar winds is therefore expected to be∼8×107K.This temperature is an order of magnitude higher than that indicated by PSPC observations;to lower it to the observed temperature of∼8.6×106K requires the mixing in of cold nebular mass that is nearly10times the mass of the combined stellar winds.This mixing may be provided by turbulent instabilities at the interface between the shocked fast winds and the cold nebular shell(e.g.,Strickland&Stevens1998)or through the hydronamic ablation of clumps of cold nebular material distributed within the hot bubble interior(Pittard,Hartquist,&Dyson2001).Assuming that mixing has taken place,we next determine the hot gas mass expected as a result of mixing and compare it to the observed value(§3.2).Given the dynamical age of the superbubble,0.13Myr,a total stellar wind mass of∼0.56M⊙has been injected to the superbubble interior,and the expected total mass of the hot gas will be5–6M⊙.This is significantly greater than the observed value of∼1M⊙.This discrepancy can be reduced if we again consider the possibility of clumpy stellar winds(Moffat&Robert1994).With the mass-loss rates reduced by a factor of≥3,the expected hot gas mass is∼2M⊙,which would be in remarkable agreement with the observed value.We summarize the comparison between the observed X-ray emission and the various models in Table3,which lists the observed X-ray luminosity and hot gas temperature and mass as well as those expected from models with and without heat conduction for both homogeneous winds and clumpy winds.It is clear that bubble models with heat conduction have the largest discrepancies from the observations.The best agreement with observed properties is from bubble models without heat conduction but allowing dynamical mixing of cold nebular material with the hot gas.For models either with or without heat conduction, clumpy winds with reduced mass loss rates are needed to minimize the discrepancy between model expectations and observations.parisons with Other Wind-Blown BubblesDiffuse X-ray emission has been previously detected from other types of wind-blown bubbles,including planetary nebulae(PNe)and circumstellar bubbles blown by Wolf-Rayet (WR)stars.The X-ray emission from these circumstellar bubbles is qualitatively and quan-titatively different from that of M17.Chu,Gruendl,&Guerrero(2003)find that the X-ray emission from PNe and WR bubbles shows a limb-brightened spatial distribution,in sharp contrast to the centrally-filled spatial distribution in M17as described in§3.1.Further,Chu et al.(2003)note that PNe and WR bubbles exhibit hot gas temperatures of1–3×106K and electron densities of10–100cm−3,while the hot interior gas of M17exhibits a temperature of8.5×106K and a substantially lower electron density of∼0.09cm−3.The comparisons of morphology and temperature between the M17superbubble and small circumstellar bubbles show fundamental differences.The limb-brightened X-ray spa-tial distribution,low temperatures,and high electron densities of PNe and WR bubbles are qualitatively consistent with a hypothesis of heat conduction and mass evaporation occurring between the hot gas interior and the swept-up shell.However,the observed X-ray luminosi-ties for PNe and WR bubbles are both significantly lower(10–100times)than predicted by bubble models with heat conduction(Chu et al.2001;Wrigge,Wendker,&Wisotzki1994; Wrigge1999).It is possible that in these wind-blown bubbles,heat conduction has also been suppressed and that dynamical mixing,which allows a lower mass injection rate,occurs at the interface between the hot gas interior and the cool nebular shell.Exploring this question will require magneticfield measurements of PNe and WR bubbles.5.SummaryWe have presented analysis of a ROSAT observation of the emission nebula M17.The blister-like morphology seen in the optical images indicates that it is a superbubble blown by the winds of its OB stars in an inhomogeneous ISM.With a stellar age of∼1Myr,M17must be a young quiescent superbubble without any supernova heating.Diffuse X-ray emission is detected from M17and is confined within the optical shell.This suggests the presence of hot106–108K gas in the interior of M17,as is expected in a wind-blown bubble.Analysis of the diffuse X-ray emission indicates a characteristic gas temperature∼8.5×106K with a mean electron number density of0.09cm−3.We have considered bubble models with and without heat conduction and found that those with heat conduction overpredict the X-ray luminosity by two orders of magnitude. Furthermore,the magneticfield measured in M17is large enough to suppress heat conduc-tion and associated mass evaporation.Bubble models without heat conduction overestimate the hot gas temperature unless mixing with cold nebular gas has occurred.If nebular gas can be dynamically mixed into the hot bubble interior and if the stellar winds are clumpy with a lower mass-loss rate,the X-ray morphology and luminosity,and hot gas temperature and mass can be reasonably reproduced.M17provides us a rare opportunity to probe the physical conditions of hot gas energized solely by shocked stellar winds in a quiescent superbubble.While we have learned much from the current analysis,our model considerations were performed on a very basic level.More robust models are needed to accurately describe the evolution of a superbubble in a medium with a large-scale density gradient,small-scale clumpiness,and a strong magneticfield.We would like to thank the anonymous referee for the stimulating comments with have helped us to improve this paper.This research has made use of data obtained through the High Energy Astrophysics Science Archive Research Center Online Service,provided by the NASA/Goddard Space Flight Center.。

Galactic aggregate 银河星集Galactic astronomy 银河系天文Galactic bar 银河系棒galactic bar 星系棒galactic cannibalism 星系吞食galactic content 星系成分galactic merge 星系并合galactic pericentre 近银心点Galactocentric distance 银心距galaxy cluster 星系团Galle ring 伽勒环Galilean transformation 伽利略变换Galileo 〈伽利略〉木星探测器gas-dust complex 气尘复合体Genesis rock 创世岩Gemini Telescope 大型双子望远镜Geoalert, Geophysical Alert Broadcast 地球物理警报广播giant granulation 巨米粒组织giant granule 巨米粒giant radio pulse 巨射电脉冲Ginga 〈星系〉X 射线天文卫星Giotto 〈乔托〉空间探测器glassceramic 微晶玻璃glitch activity 自转突变活动global change 全球变化global sensitivity 全局灵敏度GMC, giant molecular cloud 巨分子云g-mode g 模、重力模gold spot 金斑病GONG, Global Oscillation Network 太阳全球振荡监测网GroupGPS, global positioning system 全球定位系统Granat 〈石榴〉号天文卫星grand design spiral 宏象旋涡星系gravitational astronomy 引力天文gravitational lensing 引力透镜效应gravitational micro-lensing 微引力透镜效应great attractor 巨引源Great Dark Spot 大暗斑Great White Spot 大白斑grism 棱栅GRO, Gamma-Ray Observatory γ射线天文台guidscope 导星镜GW Virginis star 室女GW 型星habitable planet 可居住行星Hakucho 〈天鹅〉X 射线天文卫星Hale Telescope 海尔望远镜halo dwarf 晕族矮星halo globular cluster 晕族球状星团Hanle effect 汉勒效应hard X-ray source 硬X 射线源Hay spot 哈伊斑HEAO, High-Energy Astronomical 〈HEAO〉高能天文台Observatoryheavy-element star 重元素星heiligenschein 灵光Helene 土卫十二helicity 螺度heliocentric radial velocity 日心视向速度heliomagnetosphere 日球磁层helioseismology 日震学helium abundance 氦丰度helium main-sequence 氦主序helium-strong star 强氦线星helium white dwarf 氦白矮星Helix galaxy （NGC 2685 ）螺旋星系Herbig Ae star 赫比格Ae 型星Herbig Be star 赫比格Be 型星Herbig-Haro flow 赫比格-阿罗流Herbig-Haro shock wave 赫比格-阿罗激波hidden magnetic flux 隐磁流high-field pulsar 强磁场脉冲星highly polarized quasar （HPQ ）高偏振类星体high-mass X-ray binary 大质量X 射线双星high-metallicity cluster 高金属度星团;高金属度星系团high-resolution spectrograph 高分辨摄谱仪high-resolution spectroscopy 高分辨分光high - z 大红移Hinotori 〈火鸟〉太阳探测器Hipparcos, High Precision Parallax 〈依巴谷〉卫星Collecting SatelliteHipparcos and Tycho Catalogues 〈依巴谷〉和〈第谷〉星表holographic grating 全息光栅Hooker Telescope 胡克望远镜host galaxy 寄主星系hot R Coronae Borealis star 高温北冕R 型星HST, Hubble Space Telescope 哈勃空间望远镜Hubble age 哈勃年龄Hubble distance 哈勃距离Hubble parameter 哈勃参数Hubble velocity 哈勃速度hump cepheid 驼峰造父变星Hyad 毕团星hybrid-chromosphere star 混合色球星hybrid star 混合大气星hydrogen-deficient star 缺氢星hydrogenous atmosphere 氢型大气hypergiant 特超巨星Ida 艾达（小行星243号）IEH, International Extreme Ultraviolet 〈IEH〉国际极紫外飞行器HitchhikerIERS, International Earth Rotation 国际地球自转服务Serviceimage deconvolution 图象消旋image degradation 星象劣化image dissector 析象管image distoration 星象复原image photon counting system 成象光子计数系统image sharpening 星象增锐image spread 星象扩散度imaging polarimetry 成象偏振测量imaging spectrophotometry 成象分光光度测量immersed echelle 浸渍阶梯光栅impulsive solar flare 脉冲太阳耀斑infralateral arc 外侧晕弧infrared CCD 红外CCDinfrared corona 红外冕infrared helioseismology 红外日震学infrared index 红外infrared observatory 红外天文台infrared spectroscopy 红外分光initial earth 初始地球initial mass distribution 初始质量分布initial planet 初始行星initial star 初始恒星initial sun 初始太阳inner coma 内彗发inner halo cluster 内晕族星团integrability 可积性Integral Sign galaxy （UGC 3697 ）积分号星系integrated diode array （IDA ）集成二极管阵intensified CCD 增强CCDIntercosmos 〈国际宇宙〉天文卫星interline transfer 行间转移intermediate parent body 中间母体intermediate polar 中介偏振星international atomic time 国际原子时International Celestial Reference 国际天球参考系Frame （ICRF ）intraday variation 快速变化intranetwork element 网内元intrinsic dispersion 内廪弥散度ion spot 离子斑IPCS, Image Photon Counting System 图象光子计数器IRIS, Infrared Imager / Spectrograph 红外成象器/摄谱仪IRPS, Infrared Photometer / Spectro- 红外光度计/分光计meterirregular cluster 不规则星团; 不规则星系团IRTF, NASA Infrared Telescope 〈IRTF〉美国宇航局红外Facility 望远镜IRTS, Infrared Telescope in Space 〈IRTS〉空间红外望远镜ISO, Infrared Space Observatory 〈ISO〉红外空间天文台isochrone method 等龄线法IUE, International Ultraviolet 〈IUE〉国际紫外探测器ExplorerJewel Box （NGC 4755 ）宝盒星团Jovian magnetosphere 木星磁层Jovian ring 木星环Jovian ringlet 木星细环Jovian seismology 木震学jovicentric orbit 木心轨道J-type star J 型星Juliet 天卫十一Jupiter-crossing asteroid 越木小行星Kalman filter 卡尔曼滤波器KAO, Kuiper Air-borne Observatory 〈柯伊伯〉机载望远镜Keck ⅠTelescope 凯克Ⅰ望远镜Keck ⅡTelescope 凯克Ⅱ望远镜Kuiper belt 柯伊伯带Kuiper-belt object 柯伊伯带天体Kuiper disk 柯伊伯盘LAMOST, Large Multi-Object Fibre 大型多天体分光望远镜Spectroscopic TelescopeLaplacian plane 拉普拉斯平面late cluster 晚型星系团LBT, Large Binocular Telescope 〈LBT〉大型双筒望远镜lead oxide vidicon 氧化铅光导摄象管Leo Triplet 狮子三重星系LEST, Large Earth-based Solar 〈LEST〉大型地基太阳望远镜Telescopelevel-Ⅰcivilization Ⅰ级文明level-Ⅱcivilization Ⅱ级文明level-Ⅲcivilization Ⅲ级文明Leverrier ring 勒威耶环Liapunov characteristic number 李雅普诺夫特征数（LCN ）light crown 轻冕玻璃light echo 回光light-gathering aperture 聚光孔径light pollution 光污染light sensation 光感line image sensor 线成象敏感器line locking 线锁line-ratio method 谱线比法Liner, low ionization nuclear 低电离核区emission-line regionline spread function 线扩散函数LMT, Large Millimeter Telescope 〈LMT〉大型毫米波望远镜local galaxy 局域星系local inertial frame 局域惯性架local inertial system 局域惯性系local object 局域天体local star 局域恒星look-up table （LUT ）对照表low-mass X-ray binary 小质量X 射线双星low-metallicity cluster 低金属度星团;低金属度星系团low-resolution spectrograph 低分辨摄谱仪low-resolution spectroscopy 低分辨分光low - z 小红移luminosity mass 光度质量luminosity segregation 光度层化luminous blue variable 高光度蓝变星lunar atmosphere 月球大气lunar chiaroscuro 月相图Lunar Prospector 〈月球勘探者〉Ly-α forest 莱曼-α 森林MACHO （massive compact halo 晕族大质量致密天体object ）Magellan 〈麦哲伦〉金星探测器Magellan Telescope 〈麦哲伦〉望远镜magnetic canopy 磁蓬magnetic cataclysmic variable 磁激变变星magnetic curve 磁变曲线magnetic obliquity 磁夹角magnetic period 磁变周期magnetic phase 磁变相位magnitude range 星等范围main asteroid belt 主小行星带main-belt asteroid 主带小行星main resonance 主共振main-sequence band 主序带Mars-crossing asteroid 越火小行星Mars Pathfinder 火星探路者mass loss rate 质量损失率mass segregation 质量层化Mayall Telescope 梅奥尔望远镜Mclntosh classification 麦金托什分类McMullan camera 麦克马伦电子照相机mean motion resonance 平均运动共振membership of cluster of galaxies 星系团成员membership of star cluster 星团成员merge 并合merger 并合星系; 并合恒星merging galaxy 并合星系merging star 并合恒星mesogranulation 中米粒组织mesogranule 中米粒metallicity 金属度metallicity gradient 金属度梯度metal-poor cluster 贫金属星团metal-rich cluster 富金属星团MGS, Mars Global Surveyor 火星环球勘测者micro-arcsec astrometry 微角秒天体测量microchannel electron multiplier 微通道电子倍增管microflare 微耀斑microgravitational lens 微引力透镜microgravitational lensing 微引力透镜效应microturbulent velocity 微湍速度millimeter-wave astronomy 毫米波天文millisecond pulsar 毫秒脉冲星minimum mass 质量下限minimum variance 最小方差mixed-polarity magnetic field 极性混合磁场MMT, Multiple-Mirror Telescope 多镜面望远镜moderate-resolution spectrograph 中分辨摄谱仪moderate-resolution spectroscopy 中分辨分光modified isochrone method 改进等龄线法molecular outflow 外向分子流molecular shock 分子激波monolithic-mirror telescope 单镜面望远镜moom 行星环卫星moon-crossing asteroid 越月小行星morphological astronomy 形态天文morphology segregation 形态层化MSSSO, Mount Stromlo and Siding 斯特朗洛山和赛丁泉天文台Spring Observatorymultichannel astrometric photometer 多通道天测光度计（MAP ）multi-object spectroscopy 多天体分光multiple-arc method 复弧法multiple redshift 多重红移multiple system 多重星系multi-wavelength astronomy 多波段天文multi-wavelength astrophysics 多波段天体物。

a rXiv:as tr o-ph/14356v21May21Discovery of an X–ray pulsar in the low–mass X-ray binary 2A 1822–371Peter G.Jonker 1,Michiel van der Klis 1ABSTRACT We report the discovery of 0.59s X–ray pulsations from the low–mass X–ray binary,5.57hr dipping and eclipsing ADC source 2A 1822–371.Pulse arrival time analysis indicates a circular orbit with e <0.03(95%confidence)and an asin i for the neutron star of 1.006(5)lightseconds,implying a mass function of (2.03±0.03)×10−2M ⊙.The barycentric pulse period was 0.59325(2)s in 1996.270and 0.59308615(5)s in 1998.205,indicating an average spin up with ˙P /P =(−1.52±0.02)×10−4yr −1.For a magnetic field strength of ∼1–5×1012G as derived from the X–ray spectrum the implied intrinsic X–ray luminosity is ∼2–4×1037erg s −1.The pulse amplitude is low,but increases steeply as a function of energy from a sinusoidal amplitude of 0.25%in 2–5.4keV to ∼3%above 20keV.We discuss the constraints on the masses of the companion star and the fact that several aspects of the energy spectrum are in qualitative accordance with that of a strongly magnetised neutron star.Subject headings:accretion,accretion disks —stars:individual (2A 1822–37)—stars:neutron —stars:binaries:eclipsing —pulsars:individual (2A 1822–37)—X-rays:stars 1.Introduction The lightcurve of the low–mass X–ray binary (LMXB)2A 1822–371shows clear signs of orbital modulation in both the X–ray and optical bands (White et al.1981;Seitzer et al.1979;Masonet al.1980),with a period of 5.57hours.White et al.(1981)showed that the X–rays are emitted from an extended source,a so called Accretion Disk Corona (ADC)which is periodically partially eclipsed by the companion star (at orbital phase 0.0)as well as partially obscured by structure in the accretion disk whose height above the orbital plane varies but is greatest at phase 0.8and least at phase 0.2(White et al.1981).The implied inclination is i >70◦(Mason et al.1980).The short orbital period makes 2A 1822–371a compact LMXB.If powered by a Roche lobe filling main–sequence star the companion mass is 0.62M ⊙,however,the companion spectrum is inconsistent with that of a normal K–star (Harlaftis et al.1997).The orbital period has been measured from eclipse timing to gradually increase(Hellier et al.1990); the best ephemeris to date was provided by Parmar et al.(2001).The observed X–ray spectrum is complex and various models have been used to describe the data.With a power law index of∼1 (Parmar et al.2001)the continuum is harder than that of typical LMXBs which have power law indices of1.5–2.5.There is also evidence for a strong soft component in the1-10keV range(e.g. Heinz&Nowak2001;Parmar et al.2001).An upper limit on the presence of pulsations in the 1–30keV band of1%was derived by Hellier et al.(1992).Soon after the discovery of accreting X–ray pulsars(Giacconi et al.1971)it was realized that these are strongly magnetized(B>1012G)neutron stars accreting matter from an accretion disk(Pringle &Rees1972;Lamb et al.1973)or a stellar wind(Davidson&Ostriker1973).Whereas accretion–powered pulsars are common in massive X–ray binaries,they are rare in LMXBs,a fact that has been explained in terms of neutron star magneticfield decay(presumably accretion–induced)in the generally much longer–lived low–mass systems(Bhattacharya&Srinivasan1995).The lowerfield would allow the disk to extend to close to the neutron star and spin it up to millisecond periods. This is in accordance with binary evolutionary models predicting that LMXBs are the progenitors of binary millisecond radio pulsars(Radhakrishnan&Srinivasan1982;Alpar et al.1982;see for a detailed description Bhattacharya1995).This scenario was confirmed by the discovery of thefirst accreting millisecond pulsar,in the LMXB SAX J1808.4–3658(Wijnands&van der Klis1998). In this Letter,we report the discovery of pulsations in the LMXB2A1822–371,and describe our measurements of both the orbital Doppler shifts and the spin–up of the pulsar.We briefly discuss the constraints on the masses of the two binary components and also the energy spectrum of the pulsar.2.Observations and analysisWe used16observations obtained on1996Sept.26and27(observations1–5),1998June28and29 (observations6–11),and July24and25(observations12–16)with the proportional counter array (PCA;Jahoda et al.1996)onboard the Rossi X–ray Timing Explorer(RXTE)satellite(Bradt et al. 1993).The total amount of good data was∼73ksec.All observations yielded data with a time resolution of at least2−13s,in64energy bands covering the effective2.0–60keV energy range of RXTE.As part of a systematic search for pulsars in LMXBs(Jonker et al.in prep.),a power spectrum of Solar System barycentered data was created using an FFT technique.The Nyquist frequency was64 Hz.A weak0.59s pulsed signal was discoveredfirst in the2.0–60keV power spectrum.Investigation of the pulsed signal in various energy bands and different sub–sets of the data showed that the signal–to–noise ratio was highest in the9.4–22.7keV band of observations12–16.Therefore,we initially used this energy band and subset of the data for our analyses.We measured the Solar System barycentric pulse period in19data segments of observations12–16each with a length of∼1500s(half a typical RXTE contiguous data segment)using an epoch folding technique.The period of the pulsar showed clear evidence of the5.57hour orbital modulation due to orbital Doppler shifts with an amplitude corresponding to an asin i of about1light second. Correcting for the orbital delays using the previously known orbital ephemeris(Parmar et al.2001) and our best measure of asin i obtained from the pulse period analysis,we epoch–folded each1500s segment in observations12–16,and measured the phase of each folded profile byfitting it with a sinusoid.The residual phases were thenfitted with a model using a constant pulse period and a circular orbit.This satisfactorily described the observed dependencies on both time and orbital phase.The best–fit orbital and pulse parameters are given in Table1.The measured pulse arrival times and the best–fit orbital delay curve are displayed in Fig.1.Assuming our measured asin i and the orbital ephemeris of Parmar et al.(2001)we found for observations1–5a pulse period of0.59325(5)s.This is significantly longer than that during observations12–16(see Table1),a conclusion that is insensitive to the details of the orbital corrections.From this difference we derived a pulse period derivative of˙P=(−2.85±0.04)×10−12s s−1.Due to the weakness of the signal and the limited amounts of data we were not able to phase–connect the data within observations1–5or6–11,nor could we maintain the pulse count between the epochs of observations1–5,6–11,or12–16.Using the parameters in Table1we folded30ksec of data of observations12–16in the energy bands 2.0–5.4–9.4–13.8–22.7–60keV to measure the pulse shape and the pulse amplitude as a function of energy.The pulse profiles are consistent with being the same in each energy band and did not change significantly as a function of binary phase.The best pulse profile was obtained combining the energy bands9.4–13.8keV and13.8–22.7keV(see Fig.2).Wefitted a single sinusoid to the profile in each energy band to measure the amplitude.The pulse amplitude depends strongly on energy,increasing from0.25%±0.06%in the2.0–5.4keV band to2.8%±0.5%in the22.7–60keV band(Fig.3).The pulse amplitude was lower in observations1–5than in observations6–11and 12–16(∼1.2%versus∼1.7%and∼2.1%in the13.8–22.7keV band,respectively).Although a single sinusoid is not a perfect representation of the pulse profile this will not significantly affect the derived pulse phase differences or pulse amplitude spectrum,as the profile is constant within the errors.3.DiscussionUsing data obtained with the RXTE satellite we have discovered0.59s X–ray pulsations from the low–mass X–ray binary(LMXB)2A1822–371with˙P/P=−1.5×10−4yr−1.This is the sixth LMXB to show pulsations(Table2),the fourth whose orbital pulse delay curve was measured, and after SAX J1808.4–3658only the second compact LMXB(P orb<12hrs)for which this was done.Contrary to SAX J1808.4–3658,2A1822–371is optically bright and has a well–constrained inclination(because it is eclipsing),which might allow for a future full binary solution.Before our measurements,the nature of the compact object in2A1822–371was somewhat uncertain.HeinzFig. 1.—The arrival time delay in light seconds of the pulses due to the orbital motion of the neutron star as a function of binary phase.Phase zero is superior conjunction.Each dot represents ∼1500s of data obtained in observations 12–16.The sinusoid is the best fit to the dots.The residuals of the fit (crosses)are shown at a 10times expanded scale.Error bars are shown for the dots;for clarity they are omitted for the residuals.mean = 196.8 cts /s9.4-22.7 keVFig. 2.—The measured pulse profile obtained from epoch folding the 9.4–22.7keV data of ob-servations 12–16.The mean count rate (indicated)was subtracted.For clarity two periods are plotted.The profile is clearly non–sinusoidal.Phase zero is at HJD 2451019.4011752.The bin size is ∼0.01s.One sigma error bars are shown.&Nowak(2001)showed that it could either be a white dwarf,a neutron star or a low–mass black hole.Our detection of pulsations,together with spin period changes on a timescale of∼104years establishes that the compact object is a neutron star.We derive a mass function for the companion star of(2.03±0.03)×10−2M⊙.This,combined with the knowledge of the inclination constrains the masses of the two components to a small area in a plot of companion star mass versus neutron star mass(the shaded region in Fig.4).If the companion is a main sequence Roche–lobefilling star subject to the usual lower main sequence mass–radius relation(Kippenhahn&Wiegert1990) its mass is0.62M⊙(the horizontal line in Fig.4).This would imply a quite massive neutron star. Spectroscopic observations provide a lower limit to the semi–amplitude of the radial velocity of the companion star(Harlaftis et al.1997).From that lower limit we constrain the mass of the neutron star to be more than0.6+1.0M⊙(the vertical line in Fig.4at0.3M⊙represents the67%confidence −0.3limit).Furthermore,they showed that the inner face of the companion star is10000–15000K hotter than its back face.This is probably due to effects of X–ray heating,which could render the companion significantly undermassive for its size.For a1.4M⊙neutron star the companion has a mass of0.4–0.45M⊙(see Fig.4).An estimate of the strength of the magneticfield depends among other things on the source lumi-nosity.The luminosity of2A1822–371was estimated(Mason&Cordova1982)to be L X∼1.1×1035erg s−1(d/1kpc)2.For a distance of2.5kpc(Mason&Cordova1982)this would lead to ∼1036erg/s.However,since all observed X–rays are thought to have been scattered by an ADC (see Section1),the true source luminosity may be as high as∼1038erg/s(White&Holt1982). Such a high luminosity would be consistent with the observed binary orbital period change(Heinz &Nowak2001).From the luminosity and the spin–up rate the magneticfield can be determined (cf.Ghosh&Lamb1979);for L X∼1038erg s−1we derive a magneticfield strength,B,of∼8×1010 G,whereas for L X∼1036erg s−1B∼8×1016G,which implies that the luminosity is probably not that low.If we assume that the neutron star is spinning at its equilibrium period,then for L X∼1036erg s−1wefind B∼5×1010G and for L X∼1038erg s−1B∼5×1011G.In all this we assumed M ns=1.4M⊙,I=1045g cm2,and R=106cm for the mass,moment of inertia,and radius of the neutron star.The X–ray spectrum of2A1822–371was studied by various authors(White et al.1981;Hellier& Mason1989;Hellier et al.1992;Heinz&Nowak2001;Parmar et al.2001;Iaria et al.2001),using data obtained with different satellites(Einstein,EXOSAT,Ginga,ASCA,RXTE,and BeppoSax). Parmar et al.(2001)discussed several unusual features of the spectrum of2A1822–371and although Compton scattering in the ADC probably also affected the spectrum(White&Holt1982),in principle some of these features could be explained by the presence of a∼1012G pulsar instead of 108–109G neutron star.With a power law index of∼1the continuum spectrum is much harder than that of LMXBs of similar luminosity(Parmar et al.2001).This is,however,a common power law index for X–ray pulsars(White et al.1983).The observed cut–offat∼17keV(Parmar et al.2001)could also be explained by the presence of the pulsar.The cut–offenergy of pulsars is thought to be approximately half the cyclotron energy(Makishima&Mihara1992;see WhiteFig. 3.—Pulse amplitude as a function of energy.The horizontal bars denote the width of the energy bands,while the vertical bars denote1σuncertainties.The x–coordinate of the dots is the weighted mean photon energy in each band.The pulse amplitude increases steeply with energy..oi = 70oi = 90.Fig.4.—Companion star mass as a function of neutron star mass.The system is located between the two curves representing i=70◦and i=90◦.The region to the left of the dotted line is excluded (67%confidence)due to the lower limit on the radial velocity of the companion star(Harlaftis et al.1997).The horizontal line is the mass of the companion assuming it is a Roche–lobefilling main sequence star.The allowed region(shaded)assumes that the companion could be undermassive, not more massive than this.et al.1995for a overview).The strength of the B–field which can thus be derived from the cut–off, assuming a redshift at the neutron star surface of0.3,is∼4×1012G.The relation between the electron temperature and the energy of the cyclotron resonance(Makishima et al.1999),leads for a kT e of∼4–10keV(Parmar et al.2001;Iaria et al.2001,althoughfit with a slightly different continuum function than Makishima et al.1999)to magneticfield estimates of∼1–5×1012G, again assuming a redshift at the neutron star surface of0.3.These estimates of the magneticfield are consistent with the estimates derived from the spin–up above.The intrinsic source luminosity would be∼2–4×1037erg s−1given the˙P/P we measured.The neutron star was found to spin up on a timescale of∼6500years;˙νis(8.1±0.1)×10−12Hz s−1. Comparing this˙νwith that in the other LMXB X–ray pulsars(Table2)we note that the spin–up rate measured over∼666days is large for an LMXB X–ray pulsar,but that of the transient system GRO J1744–28is even larger(Finger et al.1996).Recent observational evidence summarized by Bildsten et al.(1997)reveals alternating episodes of spin–up and spin–down in disk–fed neutron stars.If the˙νwe measured of2A1822–371between1996.270and1998.205would be the average of multiple spin–up and spin–down episodes,then the maximum spin–up rate would be even higher. However,episodes of steady spin–up or spin–down lasting nearly a decade have been observed in GX1+4and4U1626–67(Chakrabarty et al.1997a;Chakrabarty et al.1997b).The increase in pulse amplitude with photon energy is steeper than has been found for other low–mass X–ray pulsars(4U1626–67,Rappaport et al.1977;Her X–1,White et al.1983;SAX J1808.4–3658,Wijnands&van der Klis1998).Furthermore,the pulse amplitude is low compared with other LMXB X–ray pulsars.Previous studies revealed that in2A1822–371scattering is important(White et al.1981;Hellier&Mason1989;Hellier et al.1992;Parmar et al.2001;Heinz&Nowak2001). Multiple scatterings of the pulsed emission in an ADC of0.3R⊙(White et al.1981)would have washed out the pulse due to light travel time delays.Therefore,at least a portion of the ADC should not be very optically pton scattering in such an ADC could explain the observed pulse amplitude spectrum.This work was supported in part by the Netherlands Organization for Scientific Research(NWO) grant614-51-002.This research has made use of data obtained through the High Energy Astro-physics Science Archive Research Center Online Service,provided by the NASA/Goddard Space Flight Center.This work was supported by NWO Spinoza grant08-0to E.P.J.van den Heuvel. PGJ would like to thank Rob Fender and Jeroen Homan for carefully reading an earlier version of the manuscript and Tiziana Di Salvo for discussions on the X–ray spectrum.REFERENCESAlpar,M.A.,Cheng,A.F.,Ruderman,M.A.,&Shaham,J.1982,Nature,300,728Bhattacharya, D.1995,Millisecond pulsars,eds Lewin,van Paradijs,van den Heuvel(ISBN 052141684,Cambridge University Press,1995.)Bhattacharya,D.&Srinivasan,G.1995,The magneticfields of neutron stars and their evolution, eds Lewin,van Paradijs,van den Heuvel(ISBN052141684,Cambridge University Press, 1995.)Bildsten,L.,Chakrabarty,D.,Chiu,J.,Finger,M.H.,Koh,D.T.,Nelson,R.W.,Prince,T.A., Rubin,B.C.,Scott,D.M.,Stollberg,M.,Vaughan,B.A.,Wilson,C.A.,&Wilson,R.B.1997,ApJS,113,367+Bradt,H.V.,Rothschild,R.E.,&Swank,J.H.1993,A&AS,97,355Chakrabarty,D.,Bildsten,L.,Finger,M.H.,Grunsfeld,J.M.,Koh,D.T.,Nelson,R.W.,Prince, T.A.,Vaughan,B.A.,&Wilson,R.B.1997a,ApJ,481,L101Chakrabarty,D.,Bildsten,L.,Grunsfeld,J.M.,Koh,D.T.,Prince,T.A.,Vaughan,B.A.,Finger, M.H.,Scott,D.M.,&Wilson,R.B.1997b,ApJ,474,414+Chakrabarty,D.&Morgan,E.H.1998,Nature,394,346Davidson,K.&Ostriker,J.P.1973,ApJ,179,585Finger,M.H.,Koh,D.T.,Nelson,R.W.,Prince,T.A.,Vaughan,B.A.,&Wilson,R.B.1996, Nature,381,291Ghosh,P.&Lamb,F.K.1979,ApJ,234,296Giacconi,R.,Gursky,H.,Kellogg,E.,Schreier,E.,&Tananbaum,H.1971,ApJ,167,L67 Harlaftis,E.,Charles,P.A.,&Horne,K.1997,MNRAS,285,673Heinz,S.&Nowak,M.A.2001,MNRAS,320,249Hellier,C.&Mason,K.O.1989,MNRAS,239,715Hellier,C.,Mason,K.O.,Smale,A.P.,&Kilkenny,D.1990,MNRAS,244,39Hellier,C.,Mason,K.O.,&Williams,O.R.1992,MNRAS,258,457Iaria,R.,Di Salvo,T.,Burderi,L.,&Robba,N.R.2001,ApJ,in pressJahoda,K.,Swank,J.H.,Giles,A.B.,Stark,M.J.,Strohmayer,T.,Zhang,W.,&Morgan,E.H.1996,Proc.SPIE,2808,59Kippenhahn,R.&Wiegert,A.1990,Stellar structure and evolutionLamb,F.K.,Pethick,C.J.,&Pines,D.1973,ApJ,184,271Lewin,W.H.G.,Ricker,G.R.,&McClintock,J.E.1971,ApJ,169,L17Makishima,K.&Mihara,T.1992,Frontiers in X–ray astronomy(Proc.of the28th Yamada ConferenceMakishima,K.,Mihara,T.,Nagase,F.,&Tanaka,Y.1999,ApJ,525,978Mason,K.O.&Cordova,F.A.1982,ApJ,262,253Mason,K.O.,Seitzer,P.,Tuohy,I.R.,Hunt,L.K.,Middleditch,J.,Nelson,J.E.,&White,N.E.1980,ApJ,242,L109Parmar,A.N.,Oosterbroek,T.,Del Sordo,S.,Segreto,A.,Santangelo,A.,Dal Fiume,D.,& Orlandini,M.2001,A&A,356,175Pereira,M.G.,Braga,J.,&Jablonski,F.1999,ApJ,526,L105Pringle,J.E.&Rees,M.J.1972,A&A,21,1+Radhakrishnan,V.&Srinivasan,G.1982,Curr.Sci,51,1096Rappaport,S.,Markert,T.,Li,F.K.,Clark,G.W.,Jernigan,J.G.,&McClintock,J.E.1977, ApJ,217,L29Seitzer,P.,Tuohy,I.R.,Mason,K.O.,Middleditch,J.,Nelson,J.,&White,N.E.1979,IAU Circ., 3406,1+Tananbaum,H.,Gursky,H.,Kellogg,E.M.,Levinson,R.,Schreier,E.,&Giacconi,R.1972,ApJ, 174,L143Verbunt,F.&van den Heuvel,E.1995,Formation andevolution of neutron stars and black holes in binaries,eds Lewin,van Paradijs,van den Heuvel(ISBN052141684,Cambridge University Press,1995.)White,N.,Nagase,F.,&Parmar,A.N.1995,in X-ray binaries(Cambridge Astrophysics Series, Cambridge,MA:Cambridge University Press,—c1995,edited by Lewin,Walter H.G.;Van Paradijs,Jan;Van den Heuvel,Edward P.J.),p.1White,N.E.,Becker,R.H.,Boldt,E.A.,Holt,S.S.,Serlemitsos,P.J.,&Swank,J.H.1981,ApJ, 247,994White,N.E.&Holt,S.S.1982,ApJ,257,318White,N.E.,Swank,J.H.,&Holt,S.S.1983,ApJ,270,711Wijnands,R.&van der Klis,M.1998,Nature,394,344Table1.Orbital parameters of2A1822–371.The number in brackets indicates the1σuncertainty in the last digit.Barycentric pulse period(s)at1996.270,P19960.59325(2)Pulse period derivative(s/s),˙P(−2.85±0.04)×10−12–11–paring the X–ray pulsars in LMXBs.Source P P ulse(s)˙ν(Hz/s)P orb(days)asin i References1Wijnands&van der Klis19982Chakrabarty&Morgan19983Finger et al.19964Bildsten et al.19975this paper6Tananbaum et al.19727Rappaport et al.19778Lewin et al.19719Chakrabarty et al.1997a10Pereira et al.1999。