1: % FUSE Cluster cooling flow paper, Oegerle et.al.
2: % ApJ manuscript style
3: %
4: % ----------------------------------------------------------------
5: \documentclass[12pt,preprint]{aastex}
6:
7: %\slugcomment{accepted by {\em The Astrophysical Journal}, June 14, 2001 }
8:
9: \newcommand\LL{\mbox{$\:\lambda\lambda $}}
10: \newcommand\lam{$\lambda$}
11: \newcommand{\kms}{km~s$^{-1}$}
12: \newcommand\etal{et~al.}
13: \newcommand\HI{H\,{\sc i}}
14: \newcommand\NI{N\,{\sc i}}
15: \newcommand\CII{C\,{\sc ii}}
16: \newcommand\CIII{C\,{\sc iii}}
17: \newcommand\OI{O\,{\sc i}}
18: \newcommand\OVI{O\,{\sc vi}}
19: \newcommand\OVIII{O\,{\sc viii}}
20: \newcommand\FeX{Fe\,{\sc X}}
21: \newcommand\FeIX{Fe\,{\sc IX}}
22: \newcommand\FeXXIV{Fe\,{\sc xxiv}}
23: \newcommand\FeXVII{Fe\,{\sc xvii}}
24:
25: \begin{document}
26:
27: \title{FUSE Observations of Cooling Flow Gas
28: in the Galaxy Clusters A1795 and A2597}
29:
30: \author{W. R. Oegerle\altaffilmark{1},
31: L. Cowie\altaffilmark{2},
32: A. Davidsen\altaffilmark{3}
33: E. Hu\altaffilmark{2},
34: J. Hutchings\altaffilmark{4},
35: E. Murphy\altaffilmark{5}, \\
36: K. Sembach\altaffilmark{3}
37: B. Woodgate\altaffilmark{1}
38: }
39:
40: \altaffiltext{1}{Laboratory for Astronomy and Solar Physics, Code 681,
41: Goddard Space Flight Center, Greenbelt, MD 20771}
42: \altaffiltext{2}{Institute for Astronomy, University of Hawaii,
43: Honolulu, HI 96822}
44: \altaffiltext{3}{Department of Physics and Astronomy, Johns Hopkins University,
45: Baltimore, MD 21218}
46: \altaffiltext{4}{Dominion Astrophysical Observatory, HIA/NRC of Canada,
47: Victoria, BC V9E 2E7 Canada}
48: \altaffiltext{5}{Department of Astronomy, University of Virginia,
49: Charlottsville, VA 22903}
50:
51: \begin{abstract}
52:
53: We present far-ultraviolet spectroscopy of the cores of the massive
54: cooling flow clusters Abell 1795 and 2597 obtained with FUSE. As the
55: intracluster gas cools through $3\times10^5$K, it should emit strongly
56: in the \OVI\ \LL1032,1038 resonance lines. We report the detection of
57: \OVI\ \lam1032 emission in A2597, with a line flux of $1.35 \pm 0.35
58: \times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$, as well as detection of
59: emission from \CIII\ \lam977. A marginal detection of \CIII\ \lam977
60: emission is also reported for A1795. These observations provide
61: evidence for a direct link between the hot ($10^7$ K) cooling flow gas
62: and the cool ($10^4$ K) gas in the optical emission line filaments.
63: Assuming simple cooling flow models, the \OVI\ line flux in A2597
64: corresponds to a mass deposition rate of $\sim 40 M_\odot$ yr$^{-1}$
65: within the central 36 kpc. Emission from \OVI\ \lam1032 was not
66: detected in A1795, with an upper limit of $1.5 \times 10^{-15}$ erg
67: cm$^{-2}$ s$^{-1}$, corresponding to a limit on the mass cooling flow
68: rate of $\dot{M}(28 {\rm ~kpc}) < 28 M_\odot$ yr$^{-1}$. We have
69: considered several explanations for the lack of detection of \OVI\
70: emission in A1795 and the weaker than expected flux in A2597, including
71: extinction by dust in the outer cluster, and quenching of thermal
72: conduction by magnetic fields. We conclude that a turbulent mixing
73: model, with some dust extinction, could explain our \OVI\ results while
74: also accounting for the puzzling lack of emission by \FeXVII\ in cluster
75: cooling flows.
76:
77: \end{abstract}
78:
79: \keywords{cooling flows{\em -}galaxies: clusters: individual (A1795, A2597)
80: {\em -}galaxies: intergalactic medium{\em -}ultraviolet: galaxies}
81:
82: \section{Introduction}
83:
84: X-ray observations have shown that a very hot ($\sim 10^7 - 10^8$ K)
85: diffuse intracluster medium pervades the cores of clusters of galaxies
86: \citep{fj82}. Without continued heating, this gas will cool with a
87: characteristic cooling time of $t_c \sim 3T_8^{1/2}n_3^{-1}h_{50}t_H$,
88: where $T_8$ is the gas temperature in units of $10^8$ K, $n_3$ is the
89: gas density in units of $10^{-3}$ cm$^{-3}$, $h_{50}$ is the Hubble
90: constant in units of 50 \kms Mpc$^{-1}$ and $t_H$ is the Hubble time
91: \citep{fn77}. Consequently, if gas density is high enough and the
92: temperature low enough, the characteristic cooling time will be less
93: than a Hubble time \citep{cb77, fn77}. This cooling gas should ``flow''
94: quasi-hydrostatically onto the cD/D galaxy at the bottom of the cluster
95: potential well. Typical mass flows inferred from the x-ray data are $>
96: 100$ M$_\odot$ yr$^{-1}$ within the central $\sim 100$ kpc of the
97: cluster.
98:
99: As the gas cools, it will emit strongly in various X-ray lines of highly
100: ionized Fe and O for $T \la 10^7$ K \citep{canizares88}. Emission lines
101: of \OVIII\ Ly$\alpha$ and \FeXXIV\/ arising in this hot gas have been
102: detected in the cores of clusters \citep{peterson01, tamura01}. Spatially
103: resolved x-ray spectroscopy has confirmed the presence of distributed
104: cooling gas in clusters \citep{af97}.
105: As the gas cools rapidly through the $T \sim 3 \times 10^5$ K regime, it
106: should emit strongly in certain UV resonance lines -- particularly
107: the \OVI\ \LL1032,1038 doublet \citep{ec86, voit94}. Finally, as the
108: gas cools to $T \sim 10^4$ K, it should emit strongly in the hydrogen
109: lines (H$\alpha$, Ly$\alpha$). Optical H$\alpha$ emission from cooling
110: flows has been detected toward several clusters \citep{heckman81,
111: cowie83, hu85}, \citet{hu92} has also reported extended Ly$\alpha$
112: emission from a number of clusters having H$\alpha$ emission.
113:
114: Emission lines from cooling flow gas at temperatures $\sim 1-3 \times
115: 10^5$ K have yet to be convincingly detected. \citet{dixon96} observed
116: several clusters, including A1795, with the FUV spectrograph on the {\em
117: Hopkins Ultraviolet Telescope} (HUT) in an attempt to detect emission
118: from \OVI\/. No detections were made, although in the case of the
119: strongest X-ray cooling flow cluster observed, A1795, the HUT exposure
120: time was only 2000 sec. In efforts to detect gas at temperatures of
121: $10^{5.5} - 10^{6.5}$K, a number of investigators have attempted to
122: detect the \FeX\ \lam6374 coronal line in cooling flow clusters
123: \citep{anton91, megan94, yan95}. \citet{yan95} provided the the most
124: sensitive search reported to date, and found no evidence for the line
125: and set an upper limit for the surface brightness of $10^{-18}$ erg
126: cm$^{-2}$ s$^{-1}$ arcsec$^{-2}$. Emission lines from \FeXVII\ arising
127: in gas at $T \la 5 \times 10^6$K have also not been detected in a
128: recent {\em XMM} observation of A1795 \citep{tamura01}.
129:
130: Detection of emission from gas at $\sim 3 \times 10^5$ K would
131: provide the lynchpin to
132: physically connect the previously detected hot and cool gas components
133: in the cores of clusters. We have therefore begun a program with
134: the {\em Far Ultraviolet Spectroscopic Explorer} (FUSE)
135: to search for emission from \OVI\ \LL1032,1038 in the cores of the
136: x-ray bright, cooling flow clusters A1795, A2199 and A2597. Results for
137: A1795 and A2597 are presented in this paper. FUSE has twice the effective
138: area of HUT at \OVI\ \lam1032, which, when coupled with longer
139: integration times and higher spectral resolution can provide a
140: considerable increase in sensitivity over the HUT. For example, the
141: FUSE observations reported here for A1795 are a factor of 15 more sensitive
142: than the HUT observations.
143:
144: Throughout this paper, we have used a value of $H_\circ = 50$ \kms\
145: Mpc$^{-1}$. All distances scale as the inverse of
146: $H_\circ$, while observed line luminosities and derived cooling flow
147: mass deposition rates scale as $H_\circ^{-2}$.
148:
149: \section{FUSE Observations}
150:
151: Far-ultraviolet spectra of the cores of A1795 and A2597 have been obtained
152: with the FUSE observatory. A description of the FUSE
153: instrument and its in-orbit performance is described in papers by
154: \citet{moos00} and \citet{sahnow00}. In brief, FUSE spectra cover the
155: wavelength range $905-1185$\AA\ with a spectral resolution of $\sim
156: 15-20$ \kms\ (depending on wavelength) for point sources.
157: FUSE spectra are recorded in four separate optical channels, two of
158: which employ LiF optics covering the wavelength range $\sim 1000 - 1185$\AA\
159: and two which use SiC optics covering the $905-1090$\AA\ range. Dispersed
160: light is focused onto two photon-counting microchannel plate detectors.
161:
162: All observations were carried out with the LWRS spectrograph entrance
163: aperture ($30'' \times 30''$). For extended emission which fills the
164: LWRS aperture, the instrumental line width is $\sim 100$ \kms\/, which
165: is narrower than the expected line width of cooling flow gas in the core
166: of a cluster of galaxies.
167:
168: The data were processed with the FUSE calibration software ({\it
169: calfuse}, version 1.8.7). Since the signal from the source is expected
170: to be weak, extra care was taken in subtraction of the detector
171: background. Photons events with pulse heights less than 4 or greater
172: than 15 were screened out of the data, since these events are
173: predominately background in nature. Furthermore, the standard background
174: subtraction employed by {\it calfuse}, which is adequate for bright
175: objects, was not used. Instead, the background level on each detector
176: was determined from a co-addition of all exposures for each cluster.
177: The local background in the vicinity of the expected line positions was
178: measured on the 2D spectral image at locations just above and below the
179: spectra, and were then subtracted from the extracted spectra.
180:
181: We have searched the extracted spectra for emission from \OVI\ \lam1032
182: and \CIII\ \lam977. \OVI\ \lam1032 is redshifted to 1097\AA\ and
183: 1117\AA\ in A1795 and A2597 respectively. These wavelengths fall in
184: the LiF channels for both detectors 1 and 2, and for the SiC channel in
185: detector 2 (for A1795 only), and is outside the wavelength range of the
186: SiC channel in detector 1. The combined effective area of the 2 LiF
187: channels at $\sim 1100$\AA\/, based on in-flight calibrations, is about
188: 52 cm$^{-2}$. The SiC2 channel at this wavelength has an effective
189: area of only $\sim 3$ cm$^{-2}$, and hence we have not used the SiC2
190: spectrum in our analysis. The \CIII\ \lam977 line is also redshifted to
191: a wavelength covered by both LiF and SiC channels. However, the combined
192: effective areas of the LiF channels is much greater than the SiC
193: channels, and given the fact that the signal is close to the detector
194: background, we have used only the LiF spectra from channels 1 and 2.
195:
196: Terrestrial airglow lines in the far-UV \citep{feldman00} have also
197: posed a special problem in our search for emission lines from cooling
198: flows. For example, in A1795 the \CIII\ \lam977 line is redshifted to
199: 1038.7\AA\/, which is adjacent to an \OI\ airglow line. In A2597,
200: the \CIII\ line is redshifted to 1057.4\AA\/, which is potentially
201: contaminated by a terrestrial N$_2$ emission band. Finally, in A1795,
202: the \OVI\ \lam1032 line is redshifted to coincide with a band of \NI\
203: airglow lines. We have thoroughly investigated all these possible
204: contaminations and discuss the results below.
205:
206: \subsection{A1795}
207:
208: A1795 was observed on February 23, 2000 with an integration time of
209: 37,189 seconds. The cluster has a redshift of 0.0631 \citep{oh94},
210: hence the FUSE aperture subtends $50 \times 50$ kpc at the cluster. The
211: spectrograph aperture was centered on the cD galaxy at coordinates:
212: $\alpha(2000) = 13^h 48^m 52.5^s, \delta(2000)= +26^\circ 35' 35''$.
213:
214: Any \OVI\ \lam1032 emission from A1795 would be redshifted to an
215: observed wavelength of 1097\AA\/, which is unfortunately at nearly the
216: same wavelength as a complex of \NI\ terrestrial airglow lines
217: \citep{feldman00}. These lines are strong only when looking at the
218: illuminated earth. However, they are also present in spectra taken near
219: the earth limb, and are faintly visible throughout the daylit portion of
220: the orbit. To avoid the effects of these airglow lines as much as
221: possible, we have restricted our analysis of the \OVI\ line to data
222: obtained during spacecraft orbital night, limiting the integration time
223: to 17 ksec.
224:
225: The LiF1 and LiF2 spectra of A1795 were coadded
226: and the resulting spectrum at the expected
227: wavelengths of the redshifted \OVI\ lines is shown in Figure 1, where
228: the data has been binned by 0.2\AA\ ($\sim 60$ \kms). A faint continuum
229: from the cD elliptical galaxy is detected at a level of
230: $\sim 1.5 \times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$ \AA$^{-1}$, in good
231: agreement with the UV continuum flux reported by \citep{hu92} at wavelengths
232: $\lambda > 1215$\AA\ from an IUE large aperture observation.
233: This relatively large far-UV flux from an old
234: elliptical galaxy at a redshift of 0.0631
235: indicates the presence of substantial star formation activity.
236: \citet{hu88} has estimated an ongoing star formation rate of $\sim 1$
237: M$_\odot$ yr$^{-1}$ from the IUE data. \citet{smith97} and \cite{mittaz01}
238: have reported broad-band UV emission from the center of A1795, and
239: concluded that a population of early type stars is present, which
240: presumably formed from the cooling flow.
241:
242: In Figure 1, we note possible Galactic absorption lines of H$_2$ at
243: 1094 and 1100\AA\/, which makes the spectrum appear somewhat
244: noisier than it truly is. These absorption features do not correspond
245: to any metal absorption lines at the redshift of A1795.
246: In any event, there is no obvious \OVI\
247: emission. After subtracting a continuum fit to the galaxy spectrum, we
248: compute a flux in the \lam1032 line of $6 \pm 6 \times 10^{-16}$ erg
249: cm$^{-2}$ s$^{-1}$, where we have integrated from $-300$ to $+300$
250: \kms\/ about the expected line center. The uncertainty quoted here is
251: purely a random error. There is also a systematic uncertainty in the
252: continuum placement, so we estimate a 2$\sigma$ upper limit to the flux
253: in \lam1032 line of $1.5 \times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$, or in
254: terms of the average surface brightness within the FUSE LWRS aperture,
255: $\mu$(\OVI\/)$ < 1.7 \times 10^{-18}$ erg cm$^{-2}$ s$^{-1}$
256: arcsec$^{-2}$ or equivalently, $7 \times 10^{-8}$ erg cm$^{-2}$ s$^{-1}$
257: sr$^{-1}$. This then yields an upper limit to the luminosity in the
258: \OVI\ \lam1032 line of $L_{obs}($\OVI\/$) = 2.5 \times 10^{40}$
259: erg~s$^{-1}$.
260:
261: We have have also searched the A1795 spectrum for \CIII\ \lam977
262: emission using the full 37 ksec integration taken during both orbital
263: night and day. In Figure 2, we display the $1020-1060$\AA\ spectral region
264: from that dataset. The background subtraction appears to be
265: quite good, since the \CII\ \lam1036 absorption line from the Galactic
266: interstellar medium is black at line center, as expected. A faint
267: emission line appears to be present adjacent to the \OI\ \lam1039.2
268: terrestrial airglow line, and we suggest that this is emission from
269: \CIII\ \lam977 in the cooling flow of A1795, redshifted to 1038.7\AA\/.
270: The line flux is uncertain due to the airglow contamination, but we
271: estimate a flux of $1.5\pm0.7 \times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$;
272: ie. only a $\sim 2\sigma$ detection. We also note broad absorption extending
273: from $\sim 1032$\AA\ to 1037\AA\/ in the observed frame. Absorption at
274: the two ends of this range are due to \OVI\ \lam1032 and \CII\
275: \lam1036 from the Galactic ISM. The absorption at 1034\AA\ corresponds
276: in wavelength to redshifted Ly$\gamma$ in A1795. Absorption by neutral
277: hydrogen in A1795 could, in principle, be confirmed by detecting
278: Ly$\beta$ absorption. However, Ly$\beta$ is redshifted to 1090\AA\/,
279: which is very near the edge of segment B of detector 1, where
280: calibration is problematical. This wavelength is not recorded on
281: detector 2, since it lies in the gap between segments A and B.
282:
283: \subsection{A2597}
284:
285: A2597 was observed on June 26, 2000 for a total integration time of
286: 43,032 seconds. The spectrograph aperture was centered on the cD
287: galaxy at coordinates: $\alpha(2000) = 23^h 25^m 19.7^s, \delta(2000)=
288: -12^\circ 07' 27''$. The redshift of the A2597 cD galaxy is 0.0823
289: \citep{heckman89,taylor99}, thereby shifting the \OVI\ \lam1032 line to
290: to 1116.86\AA\ in the observed frame. This is in a region of the
291: spectrum where there is no observed or expected airglow emission
292: \citep{feldman00}. We have therefore used spectra obtained during both
293: orbital day and night in our analysis of A2597. The co-added spectrum
294: from the LiF1 and LiF2 channels is shown in Figure 3, binned to
295: 0.2\AA\/. A weak, but clearly detected emission line is evident at the
296: expected position of redshifted \OVI\ \lam1032. The emission line is
297: clearly visible in both the individual LiF1 and LiF2 spectra, so it is
298: very unlikely to be due to a detector artifact. Indeed, no such
299: emission feature is observed at 1117\AA\ in the combined day and night
300: (34 ksec) observation of A1795. We therefore conclude that this weak
301: emission line is redshifted \OVI\ \lam1032 from the cooling flow nebula
302: of A2597.
303:
304: We measure a flux in the \lam1032 line of $1.3 \pm 0.35 \times 10^{-15}$
305: erg cm$^{-2}$ s$^{-1}$, where we have integrated from $-300$ to $+300$
306: \kms\/ about the expected line center. The FUSE aperture subtends $65
307: \times 65$ kpc at the cluster, resulting in an average
308: surface brightness within the FUSE LWRS aperture of $\mu$(\OVI\/)$ = 1.4
309: \times 10^{-18}$ erg cm$^{-2}$ s$^{-1}$ arcsec$^{-2}$ or equivalently,
310: $6 \times 10^{-8}$ erg cm$^{-2}$ s$^{-1}$ sr$^{-1}$. The luminosity in
311: the \OVI\ \lam1032 line is then $L_{obs}($\OVI\/$) = 3.6 \times 10^{40}$
312: erg~s$^{-1}$.
313:
314: An examination of the spectrum has also revealed an emission line at
315: 1057.4\AA\/, the
316: expected position of redshifted \CIII\ \lam977 (see Figure 4).
317: \citet{feldman00} have observed emission from terrestrial N$_2$ at this
318: wavelength when pointed at the bright earth. In order to determine if
319: the observed line in our data is terrestrial in origin, we have also
320: extracted spectra of A2597 obtained during orbital night. Although
321: the night-only data was restricted to 14 ksec, the emission line at
322: 1057\AA\ is detected quite easily. Furthermore, the flux in the
323: line is the same during the night and daylit portions of the
324: orbit. Finally, we note that there is no emission line at this wavelength
325: present in the A1795 spectra. We conclude that the evidence is strong
326: that the emission line at 1057\AA\ is not of terrestrial
327: origin, and is in fact redshifted \CIII\ \lam977 from A2597.
328: The measured line flux is $4.9 \pm 0.7 {\rm (random)}
329: \pm 1.0 {\rm (systematic)} \times 10^{-15}$
330: erg cm$^{-2}$ s$^{-1}$ measured within $\pm 300$ \kms of the expected
331: line center.
332:
333: \section{Discussion}
334:
335: The flux in the \OVI\ line is a function of the temperature of the gas,
336: its distribution in the cluster, and the mass flow rate in the cooling
337: flow. \citet{ec86} and \citet{voit94} have computed the expected
338: luminosities of ultraviolet emission lines from a gas cooling from a
339: temperature $>10^6$K to $10^4$K. The luminosity in the \OVI\ line is
340: given by
341:
342: \[
343: L({\rm OVI}) = \frac{k\dot{M}}{\mu m_H} A_O \int_{T_{min}}^{T_{max}}
344: \left( \frac{3}{2} + s \right) \chi \frac{f_{i}j({\rm OVI})}{\Lambda}dT
345: \]
346: where $\dot{M}$ is the mass flow rate, $\mu m_H$ is the mean mass per atom,
347: $A_O$ is the abundance of oxygen, $f_{i}$ is the fraction of oxygen in the
348: O$^{+5}$ ionization stage, $\chi$ is the number of particles per H atom,
349: $j($\OVI\/) is the volume emissivity in the \OVI\ line, and
350: $\Lambda$ is the radiative
351: cooling function \citep{cowie81,ec86}. The gas cools from T$_{max}$
352: down to T$_{min}$ with an evolution
353: that could be isobaric, isochoric or some combination of
354: the two; $s$ is 1 for isobaric cooling and 0 for isochoric.
355: The intracluster medium (ICM) has abundances that are typically
356: $0.1-1$ times solar \citep{mushot92, af98} which can be understood
357: in terms of the ICM being composed of
358: galactic ejecta mixed with primordial gas. Note in the
359: above equation that lower metallicity (of oxygen in the present case)
360: is offset by a similar decrease in the cooling function,
361: assuming that the abundance of oxygen is not dissimilar to other metals
362: providing the bulk of the cooling. Hence, the line luminosity is rather
363: insensitive to metallicity, as first pointed out by
364: citet{cowie81}. We also note that the assumption made here is that
365: this is the luminosity of an optically thin, extended region, with no
366: self-absorption. \citet{ec86} have calculated UV line luminosities for solar
367: abundances, with T$_{min} = 10^4$~K and T$_{max} = 10^6$~K. \citet{voit94}
368: arrive at very similar results.
369: Adopting an intermediate case between isobaric and isochoric cooling,
370: the predicted luminosity in the \OVI\ \lam1032 line is $\sim 9 \times
371: 10^{38}\dot{M}_\odot$ erg~s$^{-1}$, where $\dot{M}_\odot$ is the mass
372: cooling flow rate in solar masses per year. For an optically thin
373: plasma, the luminosity of the 1038\AA\ line is half of that of the
374: 1032\AA\ line.
375:
376: \subsection{A1795}
377:
378: Based on model prediction above, our upper limit to the observed
379: luminosity of A1795 in the \OVI\ \lam1032 line implies an upper limit
380: to the mass cooling flow rate of only $\sim 28$ M$_\odot$ yr$^{-1}$.
381:
382: \citet{edge92} determined the mass flow rates in a large sample of
383: clusters observed with the {\em Einstein Observatory} and EXOSAT, using
384: a deprojection analysis of the x-ray images. For A1795, they found a
385: mass deposition rate of $\dot{M} = 478M_\odot$ yr$^{-1}$ within a radius
386: of $230$ kpc. \citet{allen00} computed mass flow rates for a sample of
387: 30 of the most luminous X-ray clusters, by applying a similar
388: deprojection technique to ROSAT HRI images, and analyzing spatially
389: resolved spectra of the clusters obtained with ASCA. Within a cooling
390: radius of $r_{cool} = 188$ kpc, \citep{allen00} finds a spectroscopic
391: mass flow rate of $\dot{M}_S = 300M_\odot$ yr$^{-1}$. However,
392: deprojection of the ROSAT images yielded $\dot{M}_I = 449M_\odot$
393: yr$^{-1}$. A deprojection analysis including intrinsic absorption using
394: the same column densities determined from the spectral analysis of the
395: ASCA data yielded a corrected value of $\dot{M}_C = 1038M_\odot$ yr$^{-1}$.
396:
397: In the generally adopted cooling flow picture, gas begins cooling from a
398: temperature of $\sim 5 \times 10^7$K at several hundred kpc from the
399: cluster center, and cools as it flows inwards with $\dot{M}(r) \propto r$,
400: where $\dot{M}(r)$ is the integrated mass deposition rate interior
401: to $r$. For A1795, the FUSE aperture has an effective radius of $28$
402: kpc, or $0.15r_{cool}$ as determined by \citet{allen00}. Assuming a
403: mass deposition rate proportional to radius, then the x-ray mass cooling
404: rate within the FUSE aperture is $\dot{M}_X = 45-155M_\odot$ yr$^{-1}$
405: depending on whether one adopts $\dot{M}_S$, $\dot{M}_I$ or $\dot{M}_C$
406: from \citep{allen00}. Our upper limit to the mass cooling flow rate of
407: $28M_\odot$ yr$^{-1}$ from the non-detection of \OVI\ \lam1032 emission
408: is about half of that derived from the ASCA x-ray spectra, assuming no
409: absorption of the \OVI\ line.
410:
411: \subsection{A2597}
412:
413: The observed \OVI\ \lam1032 luminosity in A2597 implies a cooling flow
414: rate of $\sim 40$ M$_\odot$ yr$^{-1}$, based on the model described above.
415:
416: \citet{sarazin95} has derived a cooling rate for the ambient hot gas of
417: $\dot{M} = 327M_\odot$ yr$^{-1}$ within a radius of $r_{cool} = 130$
418: kpc from ROSAT HRI x-ray images. The effective radius of the FUSE
419: aperture at the redshift of A2597 corresponds to $46$ kpc, or
420: $0.35r_{cool}$. Scaling the mass deposition rate with radius, we then
421: find an x-ray mass cooling rate within the FUSE aperture of $\dot{M}_X =
422: 115M_\odot$ yr$^{-1}$, which is almost 3 times the cooling rate of
423: $40M_\odot$ yr$^{-1}$ inferred from the the luminosity of the \OVI\
424: line.
425:
426: \subsection{Dust Scattering}
427:
428: One possible way to diminish the observed \OVI\ emission is through
429: scattering by dust. \citet{edge99} have reported the existence of
430: significant amounts of dust in the cores of some cooling flow clusters,
431: based on submillimeter observations. \citet{hu92} has investigated the
432: Ly$\alpha$/H$\alpha$ emission line ratios in cooling flow clusters, and
433: found an average excess reddening, above the foreground extinction, of
434: $E(B-V)$ in the range $0.09-0.25$. \citet{hu92} discusses the location
435: of the dust in the cluster, and favors the idea that the bulk of the
436: dust is in the extended cluster component, instead of embedded in the
437: localized cooling flow region. An extended dust component was found to
438: be in rough agreement with estimates of cluster extinction based on
439: background galaxy counts \citep{kara69}, and is supported by the lack of
440: a correlation between $E(B-V)$ with $\dot{M}$ in cooling flow systems.
441: If this is the case, then regardless of the geometry of the cooling flow
442: region, \OVI\ emission from the core of the cluster will have to pass
443: through the dust in the outer cluster on its way to the observer.
444:
445: \citet{hu92} derived a value of $E(B-V) = 0.14_{-0.10}^{+0.08}$ for
446: A1795 (although take note of the large uncertainties, which are
447: primarily systematic). Assuming an upper limit to $E(B-V)$ of 0.22,
448: this translates into $<2.9$ magnitudes of extinction in the
449: far-ultraviolet, computed assuming a standard Galactic dust extinction curve
450: with $R_V = 3.1$. If we assume that all of this extinction arises in
451: the outer cluster and scatters the \OVI\ emission, then our observed,
452: extinction-corrected line luminosity is $L_{corr}($\OVI\/$) < 3.6
453: \times 10^{41}$ erg~s$^{-1}$. This then raises the limit to the mass
454: cooling rate to $\dot{M}_{OVI} < 400 M_\odot$ yr$^{-1}$, which does not
455: constrain the x-ray models. For A2597, \citet{hu92} quotes $E(B-V) <
456: 0.11^{+0.08}_{-0.09}$ which would result in an extinction corrected mass
457: flow rate derived from the \OVI\ observation of $\dot{M}_{OVI} \sim
458: 150^{+140}_{-100} M_\odot$ yr$^{-1}$, which is in reasonable agreement
459: with the x-ray derived mass flow rate within the FUSE aperture, given
460: the uncertainties in the actual extinction value.
461:
462: \citet{vd95} have pointed out that dust in the ICM may be unlike
463: Galactic dust, and may have different extinction properties. Indeed, it
464: is clear that extinction properties are dependent upon the past
465: processing history and environment in which the dust grains reside
466: \citep{mathis92,gordon98}. However, extinction in the far-UV is
467: dominated by small grains, and these would be subject to destruction in
468: the hot ICM by sputtering. In hot gas with density $n_H$ cm$^{-3}$, a
469: grain of size $a_\mu$ microns will be destroyed in $\sim 2
470: \times 10^6 a_\mu n_H^{-1}$ yr \citep{vd95}. So, grains smaller than
471: $1\mu$m in size would be destroyed in less than a Hubble time unless the
472: grains were near the outskirts of the cluster where $n_H < 10^{-4}$
473: cm$^{-3}$. If intracluster dust at large cluster radii existed and had an
474: extinction law like that of the Small Magellanic Cloud (ie. containing
475: small grains), then the ICM extinction could be $\sim 0.5-1$ mag
476: greater at \OVI\ than it would if an average Galactic law were assumed.
477:
478: Hot gas near the cluster center should be free of dust grains with sizes
479: $<1\mu$m due to sputtering. So, any clouds that condense out of the
480: cooling flow should also be dust-free. However, the gas surrounding the
481: central galaxies in many cooling flow clusters appears to be dusty
482: \citep{sparks89,mcnamara92}. \citet{dv93} have reported significant
483: depletion of calcium in cooling flow emission line regions, and this is
484: generally attributed to dust depletion. In the core of A1795,
485: \citet{pinkney96} have resolved a 7 kpc long dust lane, whose
486: morphology matches that of the H$\alpha$ emission. \citet{koekemoer99}
487: have also resolved an elongated dust lane southeast of the nucleus of
488: A2597, together with patches of dust near the cluster core. This dusty
489: gas may have been stripped from a gas-rich galaxy passing through the
490: cluster core, or it might be the remains of a merger of such a galaxy
491: with the central cD galaxy. Consequently, these dusty filaments may be
492: very dynamic, and may form and be destroyed continuously throughout the
493: life of the cluster. The effect of dust in the core of the cluster on
494: \OVI\ emission would likely depend on its geometry and/or filling
495: factor. If the dust were confined to optical emission line filaments
496: having a small filling factor, then the the observed
497: Ly$\alpha$/H$\alpha$ emission line ratios could be altered by this dust,
498: and yet the \OVI\ emission, which arises in the intercloud regions (or
499: at cloud interfaces), could suffer much less extinction than the optical
500: emission lines.
501:
502: \subsection{Conduction and Mixing}
503:
504: The existence of magnetic fields in intracluster gas with strengths of
505: order 1 $\mu$G have been inferred from observations of Faraday rotation
506: of radio sources seen through the ICM \citep{kim91}. Tangled magnetic
507: fields, or field lines that run perpendicular to the temperature
508: gradient, can prevent thermal conduction in an ionized gas. Recent
509: x-ray observations of two clusters, A2142 \citep{markevitch00} and A3667
510: \citep{vikhlinin01}, by {\em Chandra} have revealed what appears to be
511: central, cooler intracluster gas separated from the outer, hot gas by a
512: very sharp density discontinuity. \citet{ettori00} have pointed out
513: that this sharp temperature discontinuity could not be maintained
514: without a significant suppression of heat conduction.
515:
516: Recent {\em Chandra} x-ray images of the core of A1795 reveal an 80 kpc
517: long filament of x-ray emission extending to the SSE of the cD galaxy
518: \citep{fabian01a}. There is a remarkable morphological similarity
519: between this x-ray emission and the H$\alpha$ emission. Although the
520: nature of the filament is unclear (ie. a cooling wake, or ram pressure
521: stripped gas from a moving galaxy, etc; see \citet{fabian01a}) there
522: appears to be a direct physical relationship between the x-ray cooling
523: gas and the optical emission. Although not all of this filamentary
524: structure is within the FUSE aperture, the brightest parts of the
525: filament near the cD galaxy are within the aperture (see Figure 5). If
526: the cool clouds in the filament are small enough, they will be
527: destroyed by thermal evaporation in the hot, ambient gas, unless
528: conduction is quenched by tangled magnetic fields. The existence of the
529: filament may then indicate that conduction is ineffective in cooling the
530: hot x-ray emitting gas.
531:
532: The absence of \OVI\ emission in A1795, the rather weak \OVI\ emission
533: in A2597, and the absence of observed \FeXVII\ lines in cooling flows
534: \citep{fabian01b} could then be explained by the suppression of
535: conduction. Nevertheless, we do observe \OVI\ emission in A2597 at a
536: level consistent with expectations from x-ray observations if we correct
537: the \OVI\ flux for a reasonable amount of scattering by dust. We are
538: then left to ponder why no x-ray emission has been detected by {\em XMM} from
539: gas with temperatures less than $\la 4\times10^6$K. A possible
540: explanation is provided by turbulent mixing of hot and cool gas
541: \citep{begelman90}. \citet{loewenstein90} have pointed out that the
542: inner regions of cooling flows are likely to be highly turbulent, with
543: turbulent velocities of order $\sim 100$ \kms\/. Assuming the presence
544: of entrained cool clouds, any relative motion between the hot gas and
545: the cool clouds will excite Kelvin-Helmholtz instabilities that will
546: grow and create a mixing layer at the interface. The surface area of
547: the cool material is increased dramatically by this turbulence, and
548: allows efficient diffusion of energy between the hot and cool phases
549: \citep{begelman90,slavin93}, resulting in a very short mixing time. The
550: steady-state temperature of the mixed gas is intermediate between the
551: hot and cool gas with a value of $\xi\sqrt{T_cT_h}$ \citep{begelman90},
552: where $T_c$ and $T_h$ are, respectively, the cool and hot gas temperatures
553: and $\xi$ is a factor of order unity, and depends on the ratio of the
554: efficiencies with which hot and cool gas deposit mass and energy into
555: the mixing layer. For the case we envision in cluster cooling flows, if
556: cool gas at temperature $\sim 10^4$K and hot gas at temperature
557: $\sim 10^7$K are mixed, then the intermediate gas temperature in the
558: mixing layer will be $\sim 3\times 10^5$K. This gas is at the peak of
559: the cooling curve, and would rapidly cool and emit \OVI\ \lam1032, while
560: being too cool to emit the \FeXVII\ x-ray lines.
561:
562: \section{Summary}
563:
564: We have obtained far ultraviolet spectroscopy of the cores of the galaxy
565: clusters A1795 and A2597, in order to detect emission lines arising in
566: cooling flow gas with temperatures of $1-3\times 10^5$ K. We do not
567: detect \OVI\ \lam1032 in A1795, with a $2\sigma$ upper limit of $\sim
568: 1.5\times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$. This corresponds to an
569: upper limit to the mass deposition rate of $28 M_\odot$ yr$^{-1}$ within
570: the central 28 kpc, assuming cooling models of \citet{ec86} and
571: \citet{voit94}. We report a marginal ($\sim 2\sigma$) detection of
572: emission from \CIII\ \lam977 in A1795. We have detected \OVI\ \lam1032
573: and \CIII\ \lam977 emission in A2597 providing direct evidence linking
574: the cool gas at $10^4$K with the ambient hot, $10^7$K gas. In
575: collisional ionization equilibrium, O$^{+5}$ and C$^{+2}$ are formed at
576: temperatures of $3 \times 10^5$ K and $8 \times 10^4$ K, respectively.
577: The observed \OVI\ \lam1032 line flux in A2597 is $\sim 1.3 \pm 0.35
578: \times 10^{-15}$ erg cm$^{-2}$ s$^{-1}$ and corresponds to mass
579: deposition rate of $\dot{M}(<36 {\rm ~kpc}) \sim 40 M_\odot$ yr$^{-1}$.
580: The non-detection of \OVI\ \lam1032 in A1795 could be due to scattering
581: by dust, or to quenching of thermal conduction by intracluster magnetic
582: fields, which prevents the gas from cooling below $\sim 5\times10^6$K. A
583: turbulent mixing model, together with some dust extinction, appears to
584: adequately describe our detection of \OVI\ \lam1032 emission in A2597,
585: while also accounting for recent {\em XMM} spectrosocopic results, which
586: indicate a lack of \FeXVII\ emission, and hence a lack of cooling flow
587: gas at temperatures of $\sim 4 \times 10^6$K.
588:
589: \acknowledgments
590:
591: This work is based on data obtained for the Guaranteed Time
592: Team by the NASA-CNES-CSA FUSE mission operated by the
593: Johns Hopkins University. Financial support to U. S.
594: participants has been provided by NASA contract NAS5-32985.
595: We thank Joel Bregman and Rich Mushotzky for informative discussions
596: on cooling flows and the recent {\em XMM} results.
597:
598: \begin{thebibliography}{}
599:
600: \bibitem[Allen(2000)]{allen00} Allen, S. 2000, \mnras, 315, 269
601: \bibitem[Allen \& Fabian(1997)]{af97} Allen, S. \& Fabian, A. 1997,
602: \mnras, 286, 583
603: \bibitem[Allen \& Fabian(1998)]{af98} Allen, S. \& Fabian, A. 1998,
604: \mnras, 297, L63
605: \bibitem[Anton, Wagner \& Appenzeller(1991)]{anton91} Anton, K.,
606: Wagner, S., \& Appenzeller, I. 1991, \aap, L51
607: \bibitem[Begelman \& Fabian(1990)]{begelman90} Begelman, M. \& Fabian, A.
608: 1990, \mnras, 244, 26P
609: \bibitem[Canizares, Markert, \& Donahue(1988)]{canizares88} Canizares, C.,
610: Markert \& Donahue, M. 1988, in Cooling Flows in
611: Clusters of Galaxies, Proceedings of the NATO Advanced Research Workshop,
612: Cambridge, England, June 22-26, 1987,
613: (Dordrecht: Kluwer Academic Publishers), 63
614: \bibitem[Cowie(1981)]{cowie81} Cowie, L. L. 1981, in X-ray Astronomy
615: with the Einstein Satellite: Proceedings of the Meeting, Cambridge MA,
616: January 28-30, 1980, (Dordrecht: Reidel Publishing Co.), 227
617: \bibitem[Cowie \& Binney(1977)]{cb77} Cowie, L. \& Binney, J. 1977, \apj,
618: 215, 723
619: \bibitem[Cowie \etal(1983)]{cowie83} Cowie, L., Hu, E., Jenkins, E.,
620: \& York, D. 1983, \apj, 272, 29
621: \bibitem[Dixon \etal(1996)]{dixon96} Dixon, W. V., Hurwitz, M. \& Ferguson,
622: H. 1996, \apjl, 469, L77
623: \bibitem[Donahue \& Stocke(1994)]{megan94} Donahue, M. \& Stocke, J. 1994,
624: \apj, 422, 459
625: \bibitem[Donahue \& Voit(1993)]{dv93} Donahue, M. \& Voit, M. 1993, \apj,
626: 414, 17
627: \bibitem[Edgar \& Chevalier(1986)]{ec86} Edgar, \& Chevalier, R. 1986, \apj,
628: 310, L7
629: \bibitem[Edge, Stewart \& Fabian(1992)]{edge92} Edge, A.,
630: Stewart, G. \& Fabian, A. 1992, \mnras, 258, 177
631: \bibitem[Edge \etal(1999)]{edge99} Edge, A., Ivison, R., Smail, I., Blain, A.,
632: Kneib, J.-P. 1999, \mnras, 306, 599
633: \bibitem[Ettori \& Fabian(2000)]{ettori00} Ettori, S. \& Fabian, A. 2000,
634: \mnras, 317, L57
635: \bibitem[Fabian \etal(2001a)]{fabian01a} Fabian, A. \etal\ 2001, \mnras,
636: 321, L33
637: \bibitem[Fabian \etal(2001b)]{fabian01b} Fabian, A., Mushotzky, R., Nulsen, P.
638: \& Peterson, J. 2001, \mnras, 321, L20
639: \bibitem[Fabian \& Nulsen(1977)]{fn77} Fabian, A. \& Nulsen, P. 1977,
640: \mnras, 180, 479
641: \bibitem[Feldman \etal(2000)]{feldman00} Feldman, P., Sahnow, D., Kruk, J.,
642: Murphy, E. \& Moos, H. M. 2000, \jgr, in press
643: \bibitem[Forman \& Jones(1982)]{fj82} Forman, W. \& Jones, C. 1982, \araa, 20,
644: 547
645: \bibitem[Gordon \& Clayton(1998)]{gordon98} Gordon, K. \& Clayton, G. 1998,
646: \apj, 500, 816
647: \bibitem[Heckman(1981)]{heckman81} Heckman, T. 1981, \apjl, 251, L59
648: \bibitem[Heckman \etal(1989)]{heckman89} Heckman, T., Baum, S., van Breugel, W.,
649: \& McCarthy, P. 1989, \apj, 338, 48
650: \bibitem[Hu(1988)]{hu88} Hu, E. 1988, in Cooling Flows in Clusters
651: of Galaxies, ed. A.C. Fabian (Dordrecht: Kluwer), 73
652: \bibitem[Hu(1992)]{hu92} Hu, E. 1992, \apj, 391, 608
653: \bibitem[Hu, Cowie \& Wang(1985)]{hu85} Hu, E., Cowie, L. \& Wang 1985,
654: \apjs, 59, 447
655: \bibitem[Karachentsev \& Lipovetskii(1969)]{kara69} Karachentsev, I.
656: \& Lipovetskii, V. 1969, Soviet Astron, 12, 909
657: \bibitem[Kim, Kronberg \& Tribble(1991)]{kim91} Kim, K., Kronberg, P. \&
658: Tribble, P. 1991, \apj, 379, 80
659: \bibitem[Koekemoer \etal(1999)]{koekemoer99} Koekemoer, A., O'Dea, C.,
660: Sarazin, C., McNamara, B., Donahue, M., Voit, M., Baum, S. \& Gallimore, J.
661: 1999, \apj, 525, 621
662: \bibitem[Loewenstein \& Fabian(1990)]{loewenstein90} Loewenstein, M. \&
663: Fabian, A. 1990, \mnras, 242, 120
664: \bibitem[Markevitch \etal(2000)]{markevitch00} Markevitch, M. \etal, \apj,
665: 541, 542
666: \bibitem[Mathis \& Cardelli(1992)]{mathis92} Mathis, J. \& Cardelli, J. 1992,
667: \apj, 398, 610
668: \bibitem[McNamara \& O'Connell(1992)]{mcnamara92} McNamara. B. \& O'Connell, R.
669: 1992, \apj, 393, 579
670: \bibitem[Mittaz \etal(2001)]{mittaz01} Mittaz, J. \etal\ 2001, \aap, 365, L93
671: \bibitem[Moos \etal(2000)]{moos00} Moos, H. M. \etal\ 2000, \apjl, 538, L1
672: \bibitem[Mushotzky(1992)]{mushot92} Mushotzky, R. 1992, in Clusters and
673: Superclusters of Galaxies, ed, A. C. Fabian (Dordrecht: Kluwer), 91
674: \bibitem[Oegerle \& Hill(1994)]{oh94} Oegerle, W. \& Hill, J. 1994, \apj,
675: 107, 857
676: \bibitem[Peterson \etal(2001)]{peterson01} Peterson, \etal\ 2001,
677: \aap, 365, L104
678: \bibitem[Pinkney \etal(1996)]{pinkney96} Pinkney, J. \etal 1996, \apjl, 468, L13
679: \bibitem[Sahnow \etal(2000)]{sahnow00} Sahnow, D. S. \etal\ 2000, \apjl, 538, L7
680: \bibitem[Sarazin \etal(1995)]{sarazin95} Sarazin, C., Burns, J., Roettiger, K.
681: \& McNamara, B. 1995, \apj, 447, 559
682: \bibitem[Slavin \etal(1993)]{slavin93} Slavin, J., Shull, J.M. \& Begelman, M.
683: 1993, \apj, 407, 83
684: \bibitem[Smith \etal(1997)]{smith97} Smith, E. P., \etal\ 1997, \apj, 478, 516
685: \bibitem[Sparks \etal(1989)]{sparks89} Sparks, W., Macchetto, D. \& Golombek, D.
686: 1989, \apj, 345, 153
687: \bibitem[Tamura \etal(2001)]{tamura01} Tamura, \etal\ 2001, \aap, 365, L87
688: \bibitem[Taylor \etal(1999)]{taylor99} Taylor, G., O'Dea, C., Peck, A. \&
689: Koekemoer, A. 1999, \apjl, 512, L27
690: \bibitem[Vikhlinin, Markevitch \& Murray(2001)]{vikhlinin01} Vikhlinin, A.,
691: Markevitch, M. \& Murray, S. 2001, \apjl, 549, L47
692: \bibitem[Voit, Donahue \& Slavin(1994)]{voit94} Voit, M., Donahue, M.
693: \& Slavin, J. 1994, \apjs, 95, 87
694: \bibitem[Voit \& Donahue(1995)]{vd95} Voit, M. \& Donahue, M. 1995, \apj,
695: 452, 164
696: \bibitem[Yan \& Cohen(1995)]{yan95} Yan, L. \& Cohen, J. 1995, \apj, 454, 44
697: \end{thebibliography}
698:
699: \newpage
700:
701: \begin{figure*}
702: \plotone{f1.eps}
703: \caption{A portion of the FUSE spectrum of A1795 taken during orbital
704: night (17 ksec) binned
705: by 0.2\AA\/ ($\sim 60$ \kms\/). The spectral resolution for an
706: extended source is $\sim 0.3$ \AA\/.
707: The expected positions of the redshifted \OVI\ \LL1032,1038 lines are
708: indicated. A faint continuum from the cD galaxy is detected.
709: Terrestrial airglow lines are indicated with an Earth ($\oplus$) symbol.
710: The dashed line in this figure, and subsequent figures, is the assumed
711: continuum.}
712: \end{figure*}
713:
714: \newpage
715:
716: \begin{figure*}
717: \plotone{f2.eps}
718: \caption{A portion of the spectrum of A1795, roughly centered on redshifted
719: \CIII\ \lam977 (expected at 1038.67\AA\/). This spectrum was obtained
720: during orbital
721: day and night (37 ksec). The data are binned
722: by 0.2\AA\/. The strong emission line at 1025\AA\ is terrestrial Ly$\beta$,
723: and all the other strong emission lines at $\sim 1027$\AA\ and $1039-1041$\AA\
724: are airglow lines of \OI\/. \CII\ \lam1036 and \OVI\ \lam1032 absorption
725: from the ISM in our Galaxy are indicated. The expected position of
726: redshifted \CIII\ \lam977 in A1795 is indicated. }
727: \end{figure*}
728:
729: \newpage
730:
731: \begin{figure*}
732: \plotone{f3.eps}
733: \caption{A portion of the FUSE spectrum of A2597 binned
734: by 0.2\AA\/.
735: The expected positions of the redshifted \OVI\ \LL1032,1038 lines are
736: indicated. }
737: \end{figure*}
738:
739: \newpage
740:
741: \begin{figure*}
742: \plotone{f4.eps}
743: \caption{A portion of the FUSE spectrum of A2597 binned
744: by 0.2\AA\/. The expected position of the redshifted \CIII\ \lam977 line is
745: indicated. }
746: \end{figure*}
747:
748: \begin{figure*}
749: \plotone{f5.eps}
750: \caption{A gray-scale image of the adaptively smoothed {\it Chandra}
751: ACIS 0.3-1.5 keV
752: x-ray image from \citet{fabian01a} overlayed with the position of
753: the FUSE LWRS $30''\times30''$ spectrograph aperture.}
754: \end{figure*}
755:
756: \end{document}
757: