1: % aa.dem
2: % AA vers. 5.01, LaTeX class for Astronomy & Astrophysics
3: % demonstration file
4: % (c) Springer-Verlag HD
5: % revised by EDP Sciences
6: %-----------------------------------------------------------------------
7: %Version FPI 27062006
8: %
9: \documentclass[12pt,preprint]{aastex}
10:
11:
12:
13: %define general packages
14: %\usepackage{epsfig}
15: %\usepackage{amsmath}
16: %\usepackage{natbib}
17: %\usepackage{longtable}
18: %\usepackage{rotating}
19: %
20: %
21: \begin{document}
22: %
23: \title{CO$^+$ in M 82: A Consequence of Irradiation by X-rays}
24:
25: \author{M. Spaans} \affil{Kapteyn Astronomical Institute, P.O. Box
26: 800, 9700 AV Groningen, The Netherlands}\email{spaans@astro.rug.nl}
27:
28: \and
29:
30: \author{R. Meijerink} \affil{Astronomy Department, University of
31: California, Berkeley, CA 94720, United
32: States}\email{rowin@astro.berkeley.edu}
33:
34:
35: \begin{abstract}
36: Based on its strong CO$^+$ emission it is argued that the M 82
37: star-burst galaxy is exposed to a combination of FUV and X-ray
38: radiation. The latter is likely to be the result of the star-burst
39: superwind, which leads to diffuse thermal emission at $\sim 0.7$ keV,
40: and a compact hard, 2-10 keV, source (but not an AGN).
41: Although a photon-dominated region (FUV) component is clearly present
42: in the nucleus of M 82, and capable of forming CO$^+$, only X-ray
43: irradiated gas of density $10^3-10^5$ cm$^{-3}$ can reproduce the
44: large, $\sim (1-4)\times 10^{13}$ cm$^{-2}$, columns of CO$^+$ that
45: are observed toward the proto-typical star-burst M 82. The total X-ray
46: luminosity produced by M 82 is weak, $\sim 10^{41}$ erg s$^{-1}$,
47: but this is sufficient to drive the formation of CO$^+$.
48:
49: \end{abstract}
50:
51: \keywords{X-rays: ISM --- X-rays: galaxies --- ISM: molecules --- galaxies: starburst
52: }
53: %
54: %________________________________________________________________
55:
56: \section{Introduction}
57:
58: M 82 is a star-burst galaxy located at 3.9 Mpc (Sakai \& Madore, 1999). It has a
59: bolometric luminosity of $\sim 4\times 10^{10}$ L$_\odot$ and a large
60: body of observational data exists for this system. The observed X-ray
61: luminosity of M 82 is modest at $\sim 4\times 10^{40}$ erg s$^{-1}$
62: and ROSAT and ASCA data indicate that it has a hard, strongly absorbed
63: ($N_H\sim 10^{22}$ cm$^{-2}$) power-law component and diffuse thermal
64: contributions at $kT\approx 0.6$ and $0.3$ keV (Moran \& Lehnert 1997).
65: More recent XMM observations indicate a range of temperatures from 0.2-1.6 keV,
66: peaking at 0.7 keV, for the diffuse thermal component (Read \& Stevens 2002).
67: The thermal contributions result from the star-burst superwind
68: and the star formation itself (supernova heating of the ISM).
69: There is a compact hard (2-10 keV) X-ray component that is absorbed with a
70: photon index of 1.7 in the nucleus M 82, and that varies on a 62 days
71: period (Kaaret, Simet \& Lang 2006a,b).
72: There is a modest, $4.4\times 10^{39}$ erg
73: s$^{-1}$, 2-8 keV diffuse component (Strickland \& Heckman 2007). The work of
74: Moran \& Lehnert (1997) argues that the
75: intrinsic luminosity of M 82 in the 0.1-10 keV
76: band is about four times larger when absorption is properly taken into
77: account, yielding about $10^{41}$ erg s$^{-1}$ in total.
78: The CO$^+$ chemistry presented here is most sensitive to the softer
79: diffuse thermal component (which is largely absorbed for our model clouds),
80: so subtleties in the hard 2-10 keV band do not impact the presented results.
81: There is no clear evidence of an AGN (i.e., a buried Seyfert nucleus) in M 82.
82: The X-ray luminosity is about $\sim 0.2$\% of the total
83: FUV energy budget as measured by the FIR luminosity (Pak et al.\ 2004).
84:
85: In the center of the late-type galaxy M~82, bright (and optically
86: thin) CO$^+$ N=2-1, J=5/2-3/2 (236.06 GHz) and J=3/2-1/2 (235.79 GHz)
87: rotational emission has been observed by Fuente et al.\ (2006). Based on
88: combined CN and HCO$^+$ measurements (their Figure 2 and Table 2), the
89: latter authors argue for a clumpy photon-dominated region (PDR) model
90: where $\sim 4\times 10^5$ cm$^{-3}$ clumps are embedded in an
91: inter-clump medium and are exposed to an enhancement in the FUV radiation
92: radiation field of $G_0\sim 10^4$. This enhancement is with respect to
93: the average far-ultraviolet (6-13.6 eV) instellar radiation field in the Milky
94: Way, which enjoys a typical FUV flux of $1.6\times 10^{-3}$ erg s$^{-1}$ cm$^{-2}$.
95: It is certainly established that
96: FUV photons produced by the vigorous star formation in M 82 dominate
97: the state of the molecular gas in its center (e.g., Pak et al.\ 2004
98: and references therein for ro-vibrational H$_2$ as well as [CII] and
99: [OI] fine-structure emission), and observations of CO, HCO$^+$ and CN
100: can be explained as well by a PDR interpretation. However,
101: CO$^+$ is under-abundant by at least an order of magnitude in such PDR
102: models, compared to the observed CO$^+$ columns of
103: $\sim (1-4)\times 10^{13}$ cm$^{-2}$ (Fuente et al.\ 2006).
104:
105: In this letter, we investigate whether the large CO$^+$ columns
106: measured toward M 82 by Fuente et al.\ (2006) can be explained by the
107: irradiation of molecular gas by the modest X-ray component that M 82
108: exhibits, without violating the clear merits of PDR physics.
109:
110: \section{PDR and XDR models}
111:
112: We have constructed a set of PDR and X-ray Dominated Region (XDR)
113: models from the codes described by Meijerink \& Spaans (2005),
114: Meijerink, Spaans \& Israel (2006) and Meijerink, Spaans \& Israel (2007)
115: in which we vary
116: both the incident radiation field, density and cosmic-ray ionization rate.
117: The thermal balance
118: (with line transfer) is calculated self-consistently with the chemical
119: balance through iteration. Absorption cross sections for X-rays are
120: smaller, $\sim 1/E^3$, than for FUV photons. Therefore, PDRs show a
121: stratified structure while the changes in the chemical and thermal
122: structure in XDRs are very gradual. In XDRs, additional reactions
123: for fast electrons that ionize, excite and heat the gas are
124: included. The heating efficiency in XDRs is much higher. Since we
125: focus on galaxy centers, we have assumed the metallicity to be twice
126: Solar.
127: We take the abundance of carbon to be equal to that of oxygen,
128: since the carbon abundance increases faster than oxygen for larger
129: metallicity. The precise C:O ratio does not affect our general
130: results, with the exception of O$_2$ and H$_2$O abundances (Bergin et
131: al.\ 2000, Spaans \& van Dishoeck 2001).
132: Because the metallicity, even for a system like M 82, is generally poorly
133: known in the central regions of active star formation, we have also run
134: models with Solar metallicity. We come back to these in the discussion section.
135:
136: Our models (Meijerink et al.\ 2007) have $n=10^2-10^{6.5}$~cm$^{-3}$
137: and $G_0=10^1-10^5$ ($F_X=0.01-160$~erg~s$^{-1}$~cm$^{-2}$).
138: We adopt a standard size
139: for our model clouds of 1 or 10 pc. Our X-ray spectrum follows the spectral
140: characteristics observed for M 82, corrected
141: for extinction by the observed hydrogen column density ($\sim 10^{22}$ cm$^{-2}$).
142: This attenuation is important (factor of four) and is the reason why the total
143: X-ray luminosity is about $10^{41}$ erg s$^{-1}$ (see introduction).
144: The X-ray spectral shape matters a lot for the penetration of photons in
145: particular energy bands. Still, for the case of M 82, molecular ion formation
146: is not very sensitive to the exact spectrum. Overall most of the M 82 X-ray
147: energy is emitted below a few keV. Now, a purely
148: soft spectrum would be absorbed quickly in the outer layers of the XDR,
149: while a purely hard spectrum would penetrate all the way through. But to
150: first order it is the total energy deposition rate, basically the integral of
151: the absorbed flux, that drives the chemistry (Meijerink \& Spaans 2005).
152: Our model clouds, with
153: $N_{\rm H}\sim 10^{22}$ cm$^{-2}$ per cloud, are chosen such that they absorb the
154: bulk of the diffuse thermal X-ray flux of M 82, consistent with the significant
155: absorption that the observations indicate, and thus the dependence on the exact X-ray spectral
156: shape is modest as far as the produced column densities are concerned.
157:
158: The considered densities and fluxes are
159: representative of the conditions in M 82 in terms of PDRs, but only
160: the models with $F_X\le 10$~erg~s$^{-1}$~cm$^{-2}$ should be typical
161: of the X-ray background in M 82.
162: That is, a total X-ray luminosity of $10^{41}$ erg s$^{-1}$ yields
163: $F_X\approx 0.4$ erg~s$^{-1}$~cm$^{-2}$ for a point source at 50 pc
164: from a molecular cloud.
165: We also consider larger values for $F_X$ because the thermal X-ray emission
166: is diffuse and molecular gas will enjoy a range of distances to ambient
167: X-ray sources.
168: For example, Fuente et al.\ (2006) adopt an emission size of $6''$ for
169: CO$^+$, corresponding to a linear scale of about 100 pc at the distance to
170: M 82. Imagine then that the interstellar medium of M 82 consists
171: of a large number of $\sim$ 1-10 pc clouds. If some of these clouds are at
172: distances of about 1 pc to individual sources of X-ray radiation, then
173: only about 1\% of the total X-ray luminosity per source would already yield
174: $F_X\sim 10$ erg~s$^{-1}$~cm$^{-2}$ impinging on these clouds.
175: At the same time can the covering factor of about a hundred of these clouds
176: on the sky approach 5-50\%, and thus contribute significantly to the XDR
177: signal.
178:
179:
180: \section{Results}
181:
182: Figure \ref{columns} shows the CO$^+$ column densities for a
183: $n=10^3-10^4$~cm$^{-3}$ (10 pc cloud) and a $n=10^4-10^{6.5}$~cm$^{-3}$
184: (1 pc cloud) density range. When comparing the model results with the observations, one
185: should realize that the observed CO$^+$ column densities are actually
186: determined for gas densities around $10^5$ cm$^{-3}$ and a rotational
187: excitation temperature of about 10 K (Fuente et al.\ 2003). We also show
188: lower density models in figure \ref{columns} because the excitation of
189: CO$^+$ is quite complicated (Black 1998, St\"auber et al.\ 2006).
190: We come back to this point in the discussion.
191: In table 1 an overview is given of all the relevant chemical species and
192: their ratios.
193:
194: It is obvious that XDRs allow columns of CO$^+$ that are
195: comparable to the observed range of $\sim (1-4)\times 10^{13}$
196: cm$^{-2}$ for modest densities, while $10^5$ cm$^{-3}$ gas requires a
197: few clouds to be superposed along the line of sight, which seems very
198: reasonable. From the abundances shown in figure \ref{PDRandXDR},
199: where a comparison is made between a typical XDR and PDR model, it is
200: evident that the CO$^+$ abundance is more than an order of magnitude
201: enhanced for the same total impinging flux by energy. This is a direct
202: consequence of the more significant C$^+$+OH$\rightarrow$CO$^+$+H
203: pathway in XDRs, where large amounts of C$^+$ and OH co-exist to
204: large depths (Meijerink \& Spaans 2005). Note in this that the
205: endo-thermic O+H$_2$$\rightarrow$OH+H reaction is driven efficiently at
206: the high ($\ge 100$ K) gas temperatures that pertain in XDRs even at
207: large columns (Meijerink \& Spaans 2005), augmented by the vibrational
208: excitation of H$_2$. The XDR HOC$^+$ abundances are also much larger
209: than in PDRs and the model column densities are comparable to those of
210: CO$^+$, consistent with observations. The XDR HCO$^+$/HOC$^+$ column
211: density ratios are of the order of 20-40 when the total hydrogen
212: column density is $\le 10^{23}$ cm$^{-2}$, also consistent with
213: observations (Fuente et al.\ 2006, their table 2).
214:
215: Similarly, figure \ref{PDRandXDR} and table 1
216: show that for column densities exceeding $10^{21.5}$ cm$^{-2}$ the CO$^+$/HCO$^+$
217: column density ratio reaches values of 0.01-0.1 in XDRs, and is boosted relative to PDRs
218: for the same ambient density and total impinging flux by energy.
219: Values of 0.01-0.1 can be reached for PDRs as well, but only if the columns are
220: modest, $\le 10^{21.5}$ cm$^{-2}$. All this compares well with the
221: 4.5-6.5 mag range for individual clumps in the Fuente et al.\ model. Our
222: adopted cosmic-ray ionization rate is $5\times 10^{-15}$ comparable to
223: the Fuente et al.\ value. We find (see also Meijerink et al.\ 2006)
224: that a boost in the formation of CO$^+$ through an
225: enhanced cosmic ray ionization rate does not occur because direct
226: ionization of CO is negligible and the C$^+$ abundance is simply too
227: small beyond the radical region in PDRs to react with OH.
228:
229: Finally, in their model Fuente et al.\ (2006) require
230: about 20-40 PDR clumps of $4\times 10^5$ cm$^{-3}$ density and 7 mag
231: extinction in order to reproduce the observed CN column of
232: $\sim 10^{16}$ cm$^{-2}$. Figure \ref{PDRandXDR} shows that
233: XDRs with low impinging X-ray fluxes do not exhibit strongly enhanced
234: CN abundances (with large $F_X$ they would), but have abundances similar to
235: or somewhat smaller than PDRs. Table 1 shows
236: that our PDR model with $G_0=10^{3.5}$ and $n=10^5$ cm$^{-3}$ produces a CN
237: column of a few $\times 10^{15}$ cm$^{-2}$, and requires several clumps along
238: the line of sight, consistent with the CO$^+$ requirement at that density.
239: %We have adopted a density of $10^5$ cm$^{-3}$ for our fiducial model since
240: %this fits best with the adopted size of 1 pc for the cloud.
241: The impinging FUV flux of this PDR model is a factor of 10 below the
242: best fit model of Fuente et al.\ (2005).
243: The PDR CN/HCN column density ratios of about 4-7 are also consistent with
244: observations (Fuente et al.\ 2006, their table 2). The XDR CN/HCN column
245: density ratios are about 80-180.
246: However, the HCN abundance in an XDR is not strongly boosted at all
247: for low $F_X$ (see also Lepp \& Dalgarno 1996, their figure 3). Consequently,
248: the PDR contribution will dominate the observed HCN (as well as CN) signal
249: and no inconsistency arises.
250: Our models do not experience the bi-stability effect, where a low and a
251: high ionization phase co-exist through the interplay of H$_3^+$, S$^+$
252: and O$_2$ (e.g.\ Boger \&
253: Sternberg 2006), because gas-grain neutralization is rapid.
254:
255: %
256:
257:
258:
259: \section{Discussion}
260:
261: The CO$^+$ abundance in X-ray irradiated interstellar gas is boosted,
262: relative to the FUV irradiation case. The star-burst galaxy M 82 appears to
263: need only a modest flux of X-rays, consistent with observations, in order to
264: reproduce the observed CO$^+$ column across its nuclear disk at a density of
265: $10^3-10^5$ cm$^{-3}$. We conjecture that other star-burst galaxies may
266: experience similar effects.
267:
268: The metallicity in the central regions of M 82 is poorly known. Read \& Stevens
269: (2002) find super-solar abundances for Mg and Si, but near-solar values for
270: N, O and Fe).
271: For comparison we have run models with Solar metallicity and the same
272: density and irradiation conditions. It turns out that Solar metallicity
273: lowers the abundance of CO$^+$ by about a factor of two to three for
274: columns less than a few times $10^{22}$ cm$^{-2}$, while the difference is no
275: more than $\sim 50$\% for columns larger than a few times $10^{22}$ cm$^{-2}$.
276: This is a direct consequence of the fact that a lower metallicity causes
277: a lower absorption rate of X-rays and thus a larger total column of material
278: is needed to build up the same column of ionization driven species like
279: CO+. Since the bulk of the M 82 X-rays is absorbed in our model, i.e.\
280: we have total hydrogen columns of more than $3\times 10^{22}$ cm$^{-2}$
281: (a few clumps), the impact of metallicity variations is modest.
282: Specifically, for Solar metallcity
283: and a total hydrogen column density of $2.5\times 10^{22}$ cm$^{-2}$
284: ($\sim 2$ clumps) or $5\times 10^{22}$ cm$^{-2}$ ($\sim 4$ clumps) we find
285: CO$^+$ columns of $2\times 10^{12}$ cm$^{-2}$ (a factor 2.7 lower) or
286: $5\times 10^{12}$ cm$^{-2}$ (a factor 1.5 lower), respectively.
287:
288: It would be quite useful to have H, H$_2$ and electron collisional rate
289: coefficients for CO$^+$. Still, the situation for CO$^+$ is quite special in
290: that it is a `transient' molecule (Black 1998), for which the destruction time
291: is shorter than the time to reach collisional equilibrium. As a consequence,
292: the excited state that the formation process leaves the CO$^+$ molecule in,
293: should be an integral part of the excitation analysis because inelastic
294: collisions with H$_2$ may not be able to thermalize the CO$^+$ levels.
295: Indeed, if CO$^+$ is formed in an excited state, then a low density,
296: $n=10^3-10^4$ cm$^{-3}$, gas component as shown in figure
297: \ref{columns} and table 1 can already
298: lead to large CO$^+$ emissivities because CO$^+$ column densities increase
299: with $F_X/n$.
300:
301: \begin{acknowledgements}
302: We thank J.P.\ P\'erez Beaupuits, J.\ Mart\'\i n-Pintado and
303: S.\ Garc\'\i a-Burillo for stimulating discussions on XDR chemistry.
304: We thank P.\ Kaaret for information on recent X-ray observations of M 82.
305: \end{acknowledgements}
306:
307: \begin{thebibliography}{12}
308: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
309:
310: \bibitem[]{}Bergin, E.A., Melnick, G.J., Stauffer, J.R., Ashby, M.L.N., et al.\ 2000, ApJ, 539, L129
311:
312: \bibitem[]{}Black, J.H., 1998, in ''Chemistry and Physics of Molecules and Grains in Space. Faraday Discussions No.\ 109, p.\ 257
313:
314: \bibitem[]{}Boger, G.I. \& Sternberg, A., 2006, ApJ, 645, 314
315:
316: \bibitem[]{}Fuente, A., Rodr\'\i quez-Franco, A., Gac\'\i a-Burillo, S., Mart\'\i n-Pintado, J., \& Black, J.H., 2003, A\&A, 406, 899
317:
318: \bibitem[]{}Fuente, A., Garc\'\i a-Burillo, S., Gerin, M., Teyssier, D., Usero, A., Rizzo, J.R., \& de Vicente, P., 2005, ApJ, 619, L155
319:
320: \bibitem[]{}Fuente, A., Garc\'\i a-Burillo, S., Gerin, M., Rizzo, J.R., Usero, A., Teyssier, D., Roueff, E., \& Le Bourlot, J., 2006, ApJ, 641, L1
321: 05
322:
323: \bibitem[]{}Kaaret, P., Simet, M.G., \& Lang, C.C., 2006, ApJ, 646, 174
324:
325: \bibitem[]{}Kaaret, P., Simet, M.G., \& Lang, C.C., 2006, Science, 311, 491
326:
327: \bibitem[]{}Lehnert, M.D., Heckman, T.D., \& Weaver, K.A., 1999, ApJ, 523, 575
328:
329: \bibitem[]{}Lepp, S. \& Dalgarno, A., 1996, A\&A, 306, L21
330:
331: \bibitem[{{Meijerink} \& {Spaans}(2005)}]{Meijerink2005}
332: {Meijerink}, R. \& {Spaans}, M. 2005, \aap, 436, 397
333:
334: \bibitem[{{Meijerink} {et~al.}(2006){Meijerink}, {Spaans}, \&
335: {Israel}}]{Meijerink2006}
336: {Meijerink}, R., {Spaans}, M., \& {Israel}, F.~P. 2006, ApJ, 650, 103
337:
338: \bibitem[{{Meijerink} {et~al.}(2007){Meijerink}, {Spaans}, \&
339: {Israel}}]{Meijerink2007}Meijerink, R., Spaans, M. \& Israel, F.P., 2007, A\&A, 461, 793
340:
341: \bibitem[]{}Moran, E.C. \& Lehnert, M.D., 1997, ApJ, 478, 172
342:
343: \bibitem[]{}Pak, S., Jaffe, D.T., Stacey, G.J., Bradford, C.M., Klumpe, E.W., \& Keller, L.D., 2004, ApJ, 609, 692
344:
345: \bibitem[]{}Read, A.M. \& Stevens, I.R., 2002, MNRAS, 335, L36
346:
347: \bibitem[]{}Sakai, S. \& Madore B.F., 1999, ApJ 526, 599
348:
349: \bibitem[]{}Spaans, M. \& van Dishoeck, E.F., 2001, ApJ, 548, L217
350:
351: \bibitem[]{}St\"auber, P., Benz, A.O., J\o rgensen, J.K., van Dishoeck, E.F., Doty, S.D., van der Tak, F.F.S., 2006, astro-ph/0608393
352:
353: \bibitem[]{}Strickland, D.K. \& Heckman, T.M., 2007, ApJ, 658, 258
354:
355: \end{thebibliography}
356:
357:
358:
359:
360: \begin{table}
361: \caption{Column densities and column density ratios}
362: {\footnotesize
363: \begin{tabular}{|l|l|l|l|l|l|l|l|l|}
364: \tableline
365: $N_{\rm H}$ & N(CO$^+$) & N(HOC$^+$) & N(HCO$^+$) & N(CN) & N(HCN) & CO$^+$/HCO$^+$ & HCO$^+$/HOC$^+$ & CN/HCN \\
366: \tableline
367: \multicolumn{9}{|l|}{XDR: $n=10^5$~cm$^{-3}$ and $F_x = 5.1$~erg~s$^{-1}$~cm$^{-2}$} \\
368: \tableline
369: 1.0e22 & 3.0e12 & 3.3e12 & 4.3e13 & 1.1e15 & 6.0e12 & 0.07 & 13.2 & 181 \\
370: 2.0e22 & 4.8e12 & 5.0e12 & 1.6e14 & 2.7e15 & 2.8e13 & 0.03 & 31.5 & 95.4 \\
371: 3.0e22 & 5.7e12 & 5.9e12 & 2.7e14 & 4.7e15 & 5.9e13 & 0.02 & 46.4 & 78.9 \\
372: \tableline
373: \multicolumn{9}{|l|}{XDR: $n=10^{3.5}$~cm$^{-3}$ and $F_x = 1.6$~erg~s$^{-1}$~cm$^{-2}$} \\
374: \tableline
375: %1.0e22 & 2.7e10 & 1.8e9 & 3.6e9 & 1.9e12 & 9.0e7 & 7.4 & 2.1 & 2.1e4 \\
376: %2.0e22 & 3.1e11 & 7.9e10 & 1.7e11 & 1.4e13 & 3.4e9 & 1.8 & 2.1 & 4.2e3 \\
377: 3.0e22 & 1.2e12 & 5.7e11 & 1.5e12 & 5.2e13 & 3.2e10 & 0.8 & 2.6 & 1.6e3 \\
378: 6.0e22 & 8.3e12 & 6.9e12 & 3.7e13 & 5.1e14 & 9.4e11 & 0.2 & 5.4 & 543 \\
379: 9.1e22 & 1.8e13 & 1.5e13 & 1.3e14 & 1.5e15 & 3.8e12 & 0.14 & 8.5 & 400 \\
380: \tableline
381: \multicolumn{9}{|l|}{PDR: $n=10^5$~cm$^{-3}$, $G_0 = 10^{3.5}$ and $\zeta=5\times10^{15}$ s$^{-1}$} \\
382: \tableline
383: 1.0e22 & 1.6e10 & 1.0e10 & 2.8e14 & 2.3e15 & 3.5e14 & 5.6e-5 & 2.8e4 & 6.6 \\
384: 2.0e22 & 1.7e10 & 1.5e10 & 7.8e14 & 4.5e15 & 9.5e14 & 2.2e-5 & 5.2e4 & 4.7 \\
385: 3.0e22 & 1.9e10 & 2.0e10 & 1.3e15 & 6.7e15 & 1.6e15 & 1.4e-5 & 6.5e4 & 4.3 \\
386: \tableline
387: %\noalign{\smallskip}
388: \end{tabular}
389: }
390: \label{column_and_ratios}
391: \end{table}
392:
393: \clearpage
394:
395: \begin{figure}
396: \unitlength1cm
397: \begin{minipage}[b]{7.5cm}
398: \resizebox{7.5cm}{!}{\includegraphics*[angle=0]{f1a.ps}}
399: \end{minipage}
400: \begin{minipage}[b]{7.5cm}
401: \resizebox{7.5cm}{!}{\includegraphics*[angle=0]{f1b.ps}}
402: \end{minipage}
403: \caption{CO$^+$ column densities in XDRs as derived from the models
404: calculated in Meijerink et al.\ 2007. {\it Left:} Mid density range,
405: $n=10^3-10^4$~cm$^{-3}$, and impinging fluxes
406: of $F_X=0.16-16$~erg~s$^{-1}$~cm$^{-2}$. {\it Right:} High density range,
407: $n=10^4-10^{6.5}$~cm$^{-3}$, and impinging fluxes
408: of $F_X=1.6-160$~erg~s$^{-1}$~cm$^{-2}$.}
409: \label{columns}
410: \end{figure}
411:
412: \clearpage
413:
414: \begin{figure}
415: \unitlength1cm
416: \begin{minipage}[b]{7.5cm}
417: \resizebox{7.5cm}{!}{\includegraphics*[angle=0]{f2a.ps}}
418: \end{minipage}
419: \begin{minipage}[b]{7.5cm}
420: \resizebox{7.5cm}{!}{\includegraphics*[angle=0]{f2b.ps}}
421: \end{minipage}
422: \caption{Chemical and thermal structure of a PDR with an enhanced
423: cosmic ionization rate ($\zeta=5\times10^{-15}$~s$^{-1}$) and an XDR
424: model. Density $n=10^5$~cm$^{-3}$ and
425: $G_0=10^{3.5}$/$F_X=1.6$~ergs~s$^{-1}$~cm$^{-2}$).
426: The CO$^+$ abundance is at least an order of magnitude larger in the XDR.}
427: \label{PDRandXDR}
428: \end{figure}
429:
430: \end{document}
431: