1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %% B68 N2H+ and C18O Depletion PAPER %%
3: %% %%
4: %% EDITED: 02/10/01 EAB LADA's Revisions 02/20/02 %%
5: %% %%
6: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
7: %\documentclass{aastex}
8: %\usepackage{emulateapj5}
9: %\documentclass[12pt, preprint]{aastex}
10: \documentclass[preprint]{aastex}
11: %\documentclass[preprint2]{aastex}
12:
13: %\input {../macro.tex}
14: %\input{macro.tex}
15:
16: \def\NHthree{NH$_3$} % H3+
17: \def\Hthreep{H$_3^+$} % H3+
18: \def\NtwoHp{N$_2$H$^+$} % N2H+
19: %\def\CHthreeOH{CH$_3$OH} % CH3OH
20: \def\thCO{$^{13}$CO} % N2H+
21: \def\CeiO{C$^{18}$O} % N2H+
22: \def\CsvO{C$^{17}$O} % N2H+
23: %\def\CtfS{C$^{34}$S} % N2H+
24: \def\Ntwo{N$_2$} % N2
25: \def\Htwo{H$_2$} % N2
26: \def\av{$A_V$} % N2
27: %\newcommand{\nhtwo}{\mbox{n$_{H_{2}}$}}
28: %\newcommand{\cc}{\mbox{cm$^{-3}$}}
29:
30:
31:
32: \begin{document}
33: \setcounter{figure}{0}
34:
35: \title{ N$_2$H$^{+}$ and C$^{18}$O Depletion in a Cold Dark Cloud}
36:
37: \author{Edwin A. Bergin}
38: \affil{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street,
39: Cambridge, MA 02138} \email{ebergin@cfa.harvard.edu}
40:
41:
42: \author{Jo\~ao Alves}
43: \affil{European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748
44: Garching, Germany}
45: \email{jalves@eso.org}
46:
47: \author{Tracy L. Huard }
48: \affil{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street,
49: Cambridge, MA 02138} \email{thuard@cfa.harvard.edu}
50:
51: \author{Charles J. Lada}
52: \affil{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street,
53: Cambridge, MA 02138} \email{clada@cfa.harvard.edu}
54:
55: \slugcomment{accepted by the The Astrophysical Journal (Letters)}
56: \begin{abstract}
57: We present sensitive, high angular resolution molecular-line observations
58: of C$^{18}$O and \NtwoHp\ toward the dark globule B68. We
59: directly compare these data with the near-infrared extinction measurements of
60: \citet{alves_b68}
61: to derive the first evidence for the depletion of \NtwoHp , and by
62: inference \Ntwo , in a pre-stellar dark cloud.
63: %This evidence
64: %is found by comparison of molecular emission map to the distribution of
65: %total column density derived from near-infrared extinction techniques
66: %by \citet{alves_b68}.
67: We also find widespread C$^{18}$O depletion
68: throughout the centrally condensed core of the B68 cloud. Specifically,
69: we find the \NtwoHp\
70: emission to peak in a shell partially surrounding the peak of dust extinction.
71: Moreover, the \NtwoHp\ peaks inside
72: the much larger \CeiO\ depletion hole and has a smaller depletion zone, confirming
73: theoretical predictions.
74: These data are analyzed through a direct coupling of time dependent
75: chemical models to a radiation transfer code. This analysis highlights
76: the importance of photodissociation at cloud edges and suggests that
77: the CO abundance declines by two orders of magnitude from edge to center.
78: In contrast \NtwoHp\ declines in abundance, at minimum, by at least a factor of
79: two. Indeed it is entirely possible that both \NtwoHp\ and \Ntwo\ are
80: completely absent from the central regions of the B68 core.
81: The depletion of \NtwoHp , and its parent molecule \Ntwo , opens the possibility
82: that the centers of dense cores, prior to the formation of a star,
83: may evade detection by conventional
84: methods of probing cores using molecular emission.
85: Under these conditions H$_2$D$^+$ may be the sole viable molecular probe
86: of the innermost regions of star forming cores.
87: \end{abstract}
88:
89:
90: \keywords{ISM:dust, extinction --- ISM: abundances --- ISM: clouds --- ISM:
91: individual (B68) --- ISM: molecules --- stars: formation}
92:
93: \section{Introduction}
94:
95: Over the past decade there has been converging observational and theoretical
96: evidence of differential molecular gas phase depletions
97: that tracks the dynamical evolution of star forming cores.
98: This chemical sequence
99: begins with the depletion of sulfur-bearing molecules,
100: which have strong bonds to grain surfaces and
101: are predicted to show the largest depletion ``holes'' in their emission
102: \citep[][hereafter BL97 and C97]{bl97, charnley_dep}; indeed
103: shell-like structures are now commonly observed for such molecules
104: in low
105: mass cores \citep[e.g.,][hereafter Tf02]{kuiper_l1498, ohashi_ccs, tafalla_dep}.
106: Later in the process even volatile molecules, such as CO, deplete;
107: as clearly demonstrated by
108: several observational efforts \citep{alves_l977, kramer_ic5146,
109: caselli_l1544, jessop_l1689}.
110:
111: Besides \Htwo , the molecule predicted to be least affected by the
112: condensation process is \Ntwo , which is more volatile than CO and is
113: easily removed from grain surfaces by whatever desorption process
114: acts in molecular cores (BL97; C97). The homonuclear molecule
115: \Ntwo\ is not directly observable,
116: but its chemical daughter products, \NHthree\ and \NtwoHp ,
117: are observed to trace dense cloud cores where other species
118: such as CO and CS are depleted \citep[][Tf02]{bergin_ic5146},
119: consistent with theoretical expectations \citep[BL97, C97,][]{aikawa_dep, li_dep}.
120: %Thus theory has predicted a sequence of differential depletions from
121: %CS to CO and finally \Ntwo\ \citep[e.g.,][]{bl97, aikawa_dep, li_dep},
122: %which is borne out by observations.
123:
124: In this letter we present evidence for the depletion of both \NtwoHp\
125: and \CeiO\ in the cold pre-stellar B68 dark cloud.
126: %This is the first detection
127: %of \NtwoHp\ depletion in a dark molecular cloud core and was made possible by the
128: %direct comparison of our sensitive molecular-line observations with the recent
129: %near-infrared extinction maps of the cloud made by
130: %\citet[][hereafter ALL01]{alves_b68}.
131: The detection of \NtwoHp\ depletion opens the possibility that the cores of
132: molecular clouds may evade detection by conventional techniques which rely
133: upon molecules as probes. We discuss the implications of this result
134: for future chemical and star formation studies.
135:
136: \section{Telescope and Spectrometers}
137:
138: The J=1-0 transitions of \CeiO\ (109.78218 GHz), \CsvO\ (112.359277 GHz),
139: and \NtwoHp\ (93.17378 GHz)
140: were observed during April 2000 and 2001 using the
141: IRAM 30m telescope. \CeiO\ was mapped with Nyquist sampling,
142: whereas pointed observations of \CsvO\ were
143: conducted using the \CeiO\ map as a guide.
144: For \NtwoHp\ the inner 50$''$ of B68 was mapped with Nyquist sampling,
145: but the remainder of the map
146: was obtained with honeycomb (24$''$) spacing. The half-power beam width at 110 GHz and
147: 93 GHz is
148: 22$''$ and 26$''$, respectively.
149: The system temperatures were typically $\sim$160--190 K.
150: In April 2001 we also obtained observations of
151: the J=3-2 transition of \NtwoHp\ (279.512 GHz; $\theta_{MB} \sim 10''$).
152: Each line was observed in frequency
153: switching mode with autocorrellators as the backends providing a velocity
154: resolution of 0.053 km s$^{-1}$ (\CeiO ), 0.02 km s$^{-1}$ (\NtwoHp\ 3-2),
155: and $\sim$0.1 km s$^{-1}$ (\NtwoHp\ 1-0, \CsvO ).
156: Pointing was checked frequently on nearby point sources
157: (in April 2001 Mars was located only 9$\arcdeg$ from B68)
158: yielding an uncertainty of $\sim 3''$. The data were
159: calibrated using the standard chopper wheel method and are presented here
160: on the T$_{mb}$ scale using standard calibrations from IRAM documentation.
161:
162: %The J=1-0 transitions of \CeiO\ , \CsvO , and \NtwoHp\
163: %were observed during April 2000 and 2001 using the
164: %IRAM 30m telescope. Relevant observational parameters are provided in Table~1.
165: %\CeiO\ was mapped with Nyquist sampling
166: %and pointed observations of \CsvO\ were
167: %conducted guided by the \CeiO\ map.
168: %For \NtwoHp\ the inner 50$''$ of B68 was mapped with Nyquist sampling,
169: %but the remainder of the map
170: %was obtained with honeycomb (24$''$) spacing.
171: %In April 2001 we also obtained observations of
172: %the J=3-2 transition of \NtwoHp .
173: %Each line was observed in frequency
174: %switching mode. Frequent pointing checks were performed
175: %(in April 2001 Mars was located only 9$\arcdeg$ from B68)
176: %yielding an uncertainty of $\sim 3''$. The data were
177: %calibrated using the standard chopper wheel method and are presented here
178: %on the T$_{mb}$ scale using standard calibrations from IRAM documentation.
179:
180: \section{CO and N$_2$ Depletion}
181:
182: \subsection{Observations}
183:
184: In Figure 1a and 1b we compare the integrated intensity distribution of \CeiO\ J=1--0
185: and \NtwoHp\ J=1-0 (all hyperfine components) with
186: the map of visual extinction (\av ) obtained by \citet[][hereafter ALL01]{alves_b68}.
187: The \CeiO\ emission peaks in a partial shell-like structure with a
188: radius $\sim 50''$, approximately centered on the \av\ peak.
189: Likewise, the \NtwoHp\ emission maxima form an arc centered on
190: the \av\ peak, but lie interior to the \CeiO\ shell.
191: %Both molecules
192: %have an extension towards the southeast, with \CeiO\ showing better agreement
193: %with the \av\ map in that direction.
194: %Both \CeiO\ and \NtwoHp\ show
195: %ordered line center velocity shifts
196: %(of slightly less than a line width
197: %across the core)
198: %and line width variations throughout
199: %the map. Discussing these variations is beyond the scope of this letter and
200: %will be treated in detail in a separate publication \citep{lada_b68}.
201:
202: In Figure 2a we present a direct position-by-position
203: comparison to the visual
204: extinction data using the techniques first described in \citet{lada_ic5146}.
205: For \CeiO\ there is little scatter with
206: the main feature being a dramatic break in the linear rise in
207: CO emission with visual extinction at \av\ $\sim 8$ mag. This break is followed by a
208: slow but systematic {\em decrease} in CO emission at greater extinctions.
209: As discussed by \citet{lada_ic5146}, in the absence of
210: non-LTE excitation (as expected for \CeiO ),
211: a linear correlation between integrated intensity and \av\ corresponds to
212: an constant abundance throughout the cloud. A break in the correlation toward
213: greater extinction could be the result of either a decrease in molecular abundance
214: or high opacity saturating the molecular emission.
215: %Using the
216: %ratio of \CeiO /\CsvO\ integrated intensities at the map center we derive
217: %a \CeiO\ opacity of 1.6 (assuming $^{18}$O/$^{17}$O = 3.65).
218: However, the more optically thin \CsvO\ emission ($\tau \sim 0.4$),
219: shown in the lower right of Figure 1b, displays similar behavior and
220: does not increase for A$_V >$ 10 mag and
221: we therefore interpret the structure seen in Figure 1a and 2a as the result
222: of CO depletion. A similar conclusion, using \thCO\ and \CeiO\ data,
223: has been reached by \citet{hotzel_b68}.
224:
225: In Figure 2b we present the \NtwoHp\ J=1-0 and J=3-2 (2 detections)
226: integrated intensity as a function of
227: \av .
228: %The integrated emission from two detections of \NtwoHp\ J=3-2
229: %are also shown as a function of \av\ to the lower right.
230: In Figure 1b the
231: \NtwoHp\ emission exhibits a local minima at the location of the \av\ peak.
232: %and, thus
233: %is reminiscent of a smaller version of the CO depletion hole.
234: In Figure 2b this minima is not clearly evident, and the positions in the depression
235: are located as a cluster of points with $\int T_{mb}$dv $\sim 2.25$
236: K km s$^{-1}$ for \av\ $>$ 25 mag.
237: A definitive conclusion regarding \NtwoHp\ depletion is complicated by
238: the presence of a few points at similar cloud depths with high intensities (points
239: located as the maxima on the ``ring'' in Figure 1b). We find no difference
240: between the opacity derived from fits to the hyperfine structure from emission
241: in the ``ring'' and emission coincident with the depression.
242: %Thus, the depression is not the result of increased optical depth.
243: However, due to the known presence of excitation gradients in B68 (ALL01) and the
244: expectation of significant non-LTE excitation for \NtwoHp ,
245: simple conclusions regarding depletion cannot be drawn from Figure 2b
246: \citep[e.g.,][]{bergin_ic5146}.
247:
248: One additional notable feature in Figure 2b is the lack of significant \NtwoHp\ emission
249: for \av\ $<$ 5 mag. A similar threshold is found in IC 5146
250: by \citet{bergin_ic5146}, which can be interpreted as a decrease in the
251: \NtwoHp\ abundance due to photodestruction of \Ntwo\ by the interstellar
252: radiation field (ISRF).
253:
254: \subsection{Chemistry and Radiation Transfer}
255:
256: To determine molecular abundances as a function of cloud depth
257: we have coupled the results from chemical models
258: to a 1-D spherical Monte-Carlo radiation transfer model
259: \citep{ashby_profile}.
260: Our procedure is an iterative one in which we first estimate the
261: radial abundance profile for a given
262: species from a time-dependent chemical model in which we fix profiles of the
263: radial density, n$_{H_2}$($r$), and kinetic temperature, T$(r)$.
264: The calculated abundance profile, along with the adopted density and
265: temperature profiles, are then incorporated as input to the radiation
266: transfer model (see Tf02). The radiative transfer model determines the
267: expected profile of integrated intensity, which is convolved to the
268: appropriate angular resolution to compare directly to the observed data
269: \footnote{Similar to Tf02 for \NtwoHp , we calculate the level
270: excitation without accounting for hyperfine splitting, but include
271: the effects of the splitting
272: in the calculation of the emergent spectrum (for more discussion
273: see Tf02).}.
274: The parameters of the chemical model are
275: then adjusted until the predicted integrated intensity profile
276: matches the observed one. For example, for CO the variables in
277: this iterative solution
278: are: the magnitude of the incident local ISRF,
279: the binding energy of CO to grains (effectively the desorption rate), and time.
280:
281: We fix the radial density profile used in the chemical and radiative transfer
282: models to that
283: determined from the extinction profile (ALL01).
284: Further, we adopt as the temperature profile that derived for dust in
285: pre-stellar Bonnor-Ebert spheres by \citet{zucconi_tdust}, assuming
286: T$_{dust} =$ T$_{gas}$. However, we adjust the temperature
287: profile by a uniform reduction of 2~K in order to match the observed intensities
288: of both \NtwoHp\ transitions at the center of the cloud.
289: Although the molecular data in B68 show evidence for systematic velocity
290: gradients \citep{lada_b68}, we assume a static cloud in the radiative transfer model, which may
291: overestimate the line opacities. However, we do incorporate the observed variation in
292: the spectral line width in the model.
293: The model line width includes contributions from the thermal and turbulent widths,
294: with the latter increasing as a function of radius.
295: To be successful a model must reproduce both the dependence of $\int T$dv
296: with \av\ (i.e., Figure 2) and $\Delta$v with radius (not shown).
297:
298:
299:
300: %To place the data on the same plane as the observed quantities we
301: %derive the column density from the B68 radial density profile convolved
302: %to a resolution of 25$''$, sampled
303: %every 12$''$. The model emission is then placed onto plots shown
304: %in Figures 2a,b. To compare the predicted velocity width
305: %to the data we fit all observations with a Gaussian.\footnote{For
306: %\NtwoHp\ we used
307: %a multicomponent fit, constrained by the known hyperfine structure.}
308: %The observed FWHM were then radially averaged in 10$''$ bins.
309:
310: %We are essentially fitting the observed spectra
311: %in a method similar to that performed by Tf02.
312:
313: \subsubsection{C$^{18}$O}
314:
315: For the chemistry of CO we adopted the time dependent analytical expression given
316: by \citet{caselli_ions}. This model assumes that there
317: is no gas phase CO formation and destruction, only depletion onto grain surfaces
318: and evaporation by cosmic ray impacts (using the formalism of \citet{hasegawa_cray}).
319: In the model, CO starts
320: with an equilibrium abundance of $x$(CO) $= 8.5 \times 10^{-5}$ (relative
321: to H$_2$) and is allowed to adsorb and desorb.
322: Because we have information on the abundance at cloud edges we also account
323: for \CeiO\ photodestruction using approximations of
324: \citet{vDB88}. We further adopted the standard ISRF \citep{habing68}.
325: %\begin{equation}
326: %\[
327: %n(CO) = \frac{n(CO,\infty)}{C} + n(CO,\infty)[1 - \frac{1}{C}]exp(-\frac{t}{t_0})\;\;(1).
328: %\]
329: %\end{equation}
330: %\noindent Where $n(CO,\infty)$ is the steady state CO abundance of $x$(C0) $=$ 8.5 $\times
331: %10^{-5}$, while
332: %$C$ and $t_0$ are combinations of the various
333: %timescales involved in gaseous CO formation and destruction.
334: %These include depletion ($t_{dep}$) and cosmic ray desorption ($t_{cr}$) and
335: %the reader is referenced to \citet{caselli_ions} for more detailed
336: %descriptions ($C = 1 + \frac{t_{cr}}{t_{pd}} + \frac{t_{cr}}{t_{dep}}$;
337: %$t_0 = [t_{dep} t_{pd} + t_{cr} t_{dep} + t_{cr} t_{pd}]/t_{dep}t_{pd}t_{cr}$).
338:
339: Due to the inclusion of \CeiO\ self-shielding, the
340: model is solved by lagging the solution of the CO column behind the
341: abundance determination.
342: %Using the analytical expression we varied the time until the
343: %model reproduced the \CsvO\ column
344: %density at the map center (N(C$^{17}$0 $= 2.2 \times 10^{14}$ cm$^{-2}$;
345: %determined assuming LTE at 10 K).
346: Similar to \citet{caselli_ions} and
347: \citet{aikawa_dep} no solution is found consistent with a CO
348: binding energy, $E_b =$ 960 K. Adopting
349: $E_b =$ 1210 K we find a solution at
350: t $\sim 6-7 \times 10^4$ years\footnote{As discussed in BL97 (and references therein)
351: a CO binding energy of 960 K is for CO bound to a CO ice covered surface,
352: and 1210 K is representative
353: of CO frozen on a bare silicate surface.}.
354: The abundance profile derived from this model
355: is given in Figure 3 and the comparison to the observations is given in
356: Figure 2a. To match the observed structure at low extinction required
357: reducing the UV field by adding an additional 1 mag of extinction to the model.
358: This model provides
359: an excellent fit to the \CeiO\ data from \av\ $=$ 1 to 27 mag and suggests that
360: the \CeiO\ abundance declines by over 2 orders of magnitude from \av\ $= 1.5$
361: to the center.
362:
363: \subsubsection{N$_{2}$H$^+$}
364:
365: Because of the large number of potential gas phase reaction
366: partners for \NtwoHp\ we used the chemical model described in
367: BL97.
368: %This is a gas phase chemical model adapted to include depletion
369: %from the gas phase and cosmic ray desorption.
370: %In this model the density and extinction profile is fixed by the near-IR
371: %data and the chemistry allowed to evolve.
372: The initial chemical abundances are set to evolved
373: cloud abundances taken from \citet{aikawa_disk}, except
374: $x$(CO) $=$ 8.5 $\times 10^{-5}$ and $x$(\Ntwo ) $= 2 \times 10^{-5}$.
375: In addition we use $E_b$(CO) $= 1210$ K and $E_b$(\Ntwo ) $= 750$ K.
376: This model is run until
377: the time matched the timescale determined by the CO analysis at which point
378: the theoretical CO abundance profiles predicted by the two
379: models were compared and found to be in good agreement.
380:
381: %except at low extinction
382: %because of the difference in self-shielding between the isotopologues.
383: In this fashion the abundance of CO, a major destroyer of \NtwoHp , is fixed
384: and the primary variables that alter the \NtwoHp\ concentration are the abundance of
385: heavy metals and the binding energy
386: of its pre-cursor \Ntwo . Changing the metal ion
387: abundances will uniformly change the destruction rate of \NtwoHp\ at
388: depths $> 3$ mag (where carbon is not ionized). In contrast, raising \Ntwo\ binding
389: energy will decrease the formation rate in high density regions
390: where depletion dominates.
391:
392: No match to the data given in Figure 2b is found for a constant \NtwoHp\ abundance.
393: An abundance reduction at cloud edges is required and
394: the predicted emission does not fit the cluster of data points
395: corresponding to the emission depression discussed in \S 3.1.
396: % In addition no match is found for ``solar'' metal
397: %ion abundances.
398: If we fix the metal ion abundances to the values
399: given by \citet{williams_ions}, we can match both the J $= 1-0$ and J $= 3-2$
400: emission, provided
401: we lower the \citet{zucconi_tdust}
402: temperature profile uniformly by 2~K (as mentioned earlier)
403: and raise the \Ntwo\ binding energy to
404: E$_b$(\Ntwo ) $= 900$ K.
405:
406: The ``best'' match abundance profiles for B68 are given in Figure 3 and the
407: fit to the observed emission is shown as a solid curve in Figure 2b. The \NtwoHp\
408: abundance profile includes a reduction at cloud edges,
409: peaks at \av\ $= 2$ mag with $x$(\NtwoHp ) $= 6 \times 10^{-11}$,
410: and declines by a factor of two by \av\ $ = 17$ mag.
411: Over this range the concentration of \Ntwo\ declines by a factor of 5.
412: {\em The abundance profile, and N$_2$H$^+$ depletion, is constrained by
413: the observations and the radiative transfer;} however, the parameters of the
414: chemical model may not be unique. For example,
415: the \NtwoHp\ abundance is sensitive to
416: E$_b$(\Ntwo ) and the ionization fraction.
417: Observations of other molecular ions, which we are
418: in the process of obtaining, will better constrain the ionization fraction
419: and the binding energy.
420:
421: %This time is essentially the ``depletion'' time
422: %required for CO to freeze out in an evolved cloud
423: %(e.g. one that evolved at lower density without depletion
424: %and instantaneously reaches the observed density profile without chemical
425: %changes).
426:
427: \section{Discussion}
428:
429: Our observations and analysis provide compelling evidence for
430: the existence of CO depletion in B68.
431: Perhaps the most interesting result of our experiment, however,
432: is the finding of \NtwoHp\ depletion.
433: The robustness of this result depends on the validity of our assumptions.
434: %and the precise \NtwoHp\ abundances may not be uniquely
435: %prescribed by our radiative transfer model.
436: The density profile is well constrained by observations, but the temperature
437: structure is not. B68 is a pre-stellar core; hence the most
438: likely dependence is the
439: adopted one with hot edges and a cold center.
440: Moreover, the detection of \NtwoHp\ J $= 3-2$ in the core center constrains
441: the temperature in the region where depletion is dominant.
442: Opacity is a concern. The Monte-Carlo radiation transfer
443: properly handles the photon propagation, but we have not
444: accounted for the hyperfine structure or systematic line center velocity motions.
445: However, these effects would reduce the opacity, and re-enforce
446: the claim of depletion. Consequently the evidence for \NtwoHp\ depletion
447: in B68 is relatively firm.
448: In our best fit model we have chosen to fit between the scatter
449: of points at \av\ $> 20$ mag and not the points in the emission
450: depression. Thus we likely have underestimated the \NtwoHp\ and
451: \Ntwo\ depletion. Indeed it is entirely possible that \Ntwo\ is almost completely frozen
452: out in the dense center of B68, similar to CO.
453: %Alternately if B68 is
454: %currently condensing, the depletion of \Ntwo\ will only increase.
455:
456: Depletion has been found to characterize the inner regions of many quiescent
457: cloud cores. The chemical daughter
458: products of \Ntwo\ (\NtwoHp\ and \NHthree ) have previously been identified as the
459: best available tracers of these regions.
460: However, our finding of \NtwoHp\ depletion
461: opens the possibility that the very centers of dense cores
462: could be effectively opaque to conventional methods of observation.
463: One casualty of this unfortunate situation is the search for infall wings
464: using optically thin molecular emission \citep{rawlings_dep}
465: will be hampered by depletion in the very regions where the infall speed is highest.
466: However, because \Htwo\
467: does not deplete onto grains, its chemical daughter
468: product, H$_2$D$^{+}$, could therefore prove to be the best tracer of gas in the innermost
469: regions of pre-stellar molecular cores.
470: Indeed, because of \Ntwo\ and CO depletion,
471: the abundance of H$_2$D$^{+}$ is expected to be large \citep{brown_deut}.
472: %As instruments are developed with greater sensitivity and
473: %new atmospheric windows are opened, H$_2$D$^{+}$ may become the best
474: %tracer in the use of molecular emission to probe the dynamics, chemistry,
475: %and physics of star formation.
476:
477: Of course, B68 could be a unique object and may be more stable than cores like L1544.
478: This would allow for chemistry to evolve over longer times, resulting in
479: greater freeze out. Alternatively, \citet{charnley_n2} showed that selective
480: depletion could lead to \Ntwo\ destruction by gas-phase chemistry, primarily
481: at longer timescales and higher densities.
482: Because of the potential implications of near complete
483: molecular freeze-out, additional searches for
484: \NtwoHp\ and \NHthree\
485: depletion are needed to infer the general applicability of this result.
486: Regardless, our observations and analysis provide important constraints on the
487: chemical evolutionary processes active in the B68 cloud.
488:
489:
490:
491: \acknowledgements
492:
493: We are grateful to the referee, Steve Charnley, for helpful comments.
494: This work was supported by Grant \#NAG5-9520 from NASA's Origins Program.
495:
496:
497:
498: \begin{thebibliography}{}
499:
500: \bibitem[Aikawa et al.(2001)]{aikawa_dep} Aikawa, Y., Ohashi, N., Inutsuka, S.,
501: Herbst, E., \& Takakuwa, S.\ 2001, \apj, 552, 639
502: %
503: \bibitem[Aikawa et al.(1996)]{aikawa_disk} Aikawa, Y.,
504: Miyama, S.~M., Nakano, T., \& Umebayashi, T.\ 1996, \apj, 467, 684
505: %
506: %
507: \bibitem[Alves, Lada, \& Lada (2001)]{alves_b68} Alves, J., Lada, C.~J., \&
508: Lada, E.~A.\ 2001, \nat, 409, 159 (ALL01)
509: %
510:
511: \bibitem[Alves et al.(1998)]{alves_l977} Alves, J., Lada, C.~J., Lada,
512: E.~A., Kenyon, S.~J., \& Phelps, R.\ 1998, \apj, 506, 292
513:
514: \bibitem[Ashby et al.(2000)]{ashby_profile} Ashby, M.~L.~N.~et al.\ 2000, \apjl,
515: 539, L115
516: %
517: \bibitem[Bergin \& Langer (1997)]{bl97} Bergin, E. A. \& Langer, W. D.
518: 1997, ApJ, 486, 316 (BL97)
519: %
520: \bibitem[Bergin et al.(2001)]{bergin_ic5146} Bergin, E.~A., Ciardi, D.~R., Lada,
521: C.~J., Alves, J., \& Lada, E.~A.\ 2001, \apj, 557, 209
522: %
523:
524: \bibitem[Brown \& Millar(1989)]{brown_deut} Brown, P.~D.~\&
525: Millar, T.~J.\ 1989, \mnras, 237, 661.
526:
527: \bibitem[Caselli et al.(2002)]{caselli_ions} Caselli, P., Walmsley, C.~M.,
528: Zucconi, A., Tafalla, M., Dore, L., \& Myers, P.~C.\ 2002, \apj, 565, 344
529: %
530: \bibitem[Caselli et al.(1999)]{caselli_l1544} Caselli, P., Walmsley, C.~M.,
531: Tafalla, M., Dore, L., \& Myers, P.~C.\ 1999, \apjl, 523, L165
532: %
533: \bibitem[Charnley \& Rodgers (2001)]{charnley_n2} Charnley, S.~B.\ \&
534: Rodgers, S.D. 2001, \apj , 569
535:
536: \bibitem[Charnley (1997)]{charnley_dep} Charnley, S.~B.\ 1997, \mnras, 291, 455 (C97)
537: %
538: \bibitem[Habing (1968)]{habing68} Habing, H. J. 1968,
539: Bull. Astron. Inst. Netherlands, 19, 421
540: %
541: \bibitem[Hasegawa \& Herbst (1993)]{hasegawa_cray} Hasegawa, T.~I.~\& Herbst, E.\
542: 1993, \mnras, 261, 83
543: %
544: \bibitem[Hotzel et al.(2002)]{hotzel_b68} Hotzel, S., Harju, J., Juvela, M.,
545: Mattila, K., \& Haikala, L.K. 2002, \aap , submitted
546:
547: \bibitem[Jessop \& Ward-Thompson (2001)]{jessop_l1689} Jessop, N.~E.~\&
548: Ward-Thompson, D.\ 2001, \mnras, 323, 1025
549: %
550: \bibitem[Kramer et al.(1999)]{kramer_ic5146} Kramer, C. Alves, J., Lada, C.,
551: Lada, E., Sievers, A., Ungerechts, H., \& Walmsley, M. 1999, \aap, 342,
552: 257
553:
554: \bibitem[Kuiper, Langer, \& Velusamy (1996)]{kuiper_l1498} Kuiper,
555: T. B. H., Langer, W. D. \& Velusamy, T. 1996, \apj, 468, 761
556:
557: \bibitem[Lada et al.(2002)]{lada_b68} Lada, C. J., Bergin, E.A., Alves, J.A.,
558: \& Huard, T. 2002, in preparation
559:
560: \bibitem[Lada et al.(1994)]{lada_ic5146} Lada, C. J., Lada, E. A., Clemens, D.
561: P., \& Bally, J. 1994, \apj, 429, 694
562:
563: \bibitem[Li et al.(2002)]{li_dep} Li, Z., Shematovich, V.I., Wiebe, D.S.,
564: Shustov, B.M. (2002), ApJ, in press
565:
566: \bibitem[Ohashi (2000)]{ohashi_ccs} Ohashi, N.\ 2000, Astrochemistry: From
567: Molecular Clouds to Planetary, I.A.U. 197, 61
568:
569: \bibitem[Rawlings et al.(1992)]{rawlings_dep} Rawlings,
570: J.~M.~C., Hartquist, T.~W., Menten, K.~M., \& Williams, D.~A.\ 1992, \mnras, 255, 471
571: %
572: %\bibitem[Stark, van der Tak, \& van Dishoeck (1999)]{stark_h2dp} Stark, R., van
573: %der Tak, F.~F.~S., \& van Dishoeck, E.~F.\ 1999, \apjl, 521, L67
574:
575: \bibitem[Tafalla et al.(2002)]{tafalla_dep} Tafalla, M., Myers, P.C.,
576: Caselli, P., Walmsley, C.M., \& Comito, C. 2002, ApJ, in press (Taf02)
577:
578: \bibitem[van Dishoeck \& Black (1988)]{vDB88} van Dishoeck, E.~F.~\& Black,
579: J.~H.\ 1988, \apj, 334, 771
580:
581: \bibitem[Williams et al.(1998)]{williams_ions} Williams, J.~P., Bergin, E.~A.,
582: Caselli, P., Myers, P.~C., \& Plume, R.\ 1998, \apj, 503, 689
583:
584:
585: \bibitem[Zucconi, Walmsley, \& Galli (2001)]{zucconi_tdust} Zucconi, A., Walmsley,
586: C.~M., \& Galli, D.\ 2001, \aap, 376, 650
587: %
588: %
589: \end{thebibliography}
590:
591: %\begin{deluxetable}{lllccc}
592: %\tablenum{1}
593: %\tablewidth{5in}
594: %\tablecolumns{7}
595: %\tablecaption{Observed Transitions and Telescope Parameters}
596: %\tablehead{
597: %\colhead{Molecule} &
598: %\colhead{Transition } &
599: %\colhead{$\nu$(GHz)} &
600: %\colhead{$\theta_{MB}$($''$)} &
601: %\colhead{$\Delta v_{res}$ (km s$^{-1}$)} &
602: %\colhead{$\sigma$ (K)} \\
603: %}
604: %\startdata
605: %C$^{18}$O & J $=$ 1 -- 0 & 109.78218 & 22 & 0.053 & 0.14 \\
606: %C$^{17}$O & J $=$ 1 -- 0 & 112.35928 & 21 & 0.104 & 0.05 \\
607: %N$_2$H$^+$ &J $=$ 1 -- 0 & 93.17378 & 26 & 0.126 & 0.08 \\
608: %N$_2$H$^+$ & J $=$ 3 -- 2 & 279.51194 & 10 & 0.021\tablenotemark{a} & 0.08 \\
609: %\enddata
610: %\tablenotetext{a}{To improve the signal to noise these data were smoothed to
611: %$\Delta v_{res} = 0.168$ km s$^{-1}$.}
612: %\end{deluxetable}
613: %
614:
615: \begin{center}
616: \begin{figure*}
617: \includegraphics[height=8cm]{f1.eps}
618: \caption{
619: (a) Comparison of \CeiO\ J=1-0 integrated emission (contours) in B68 superimposed on a
620: map of visual extinction derived by ALL01 and convolved to the IRAM resolution.
621: The \CeiO\ contours
622: begin at 0.2 K km s$^{-1}$ and step in units of 0.1 K km s$^{-1}$. The
623: A$_V$ image scales is scaled from 0--27 mag. (b) Comparison of
624: \NtwoHp\ J=1-0 integrated emission (contours) and the
625: visual extinction image. The \NtwoHp\ contours begin at 0.3 K km s$^{-1}$
626: and step in units of 0.2 K km s$^{-1}$.
627: These maps were referenced to $\alpha$ = 17$^h$22$^m$38\fs 2
628: and $\delta = -$23$\arcdeg$49$'$34\farcs 0 (J2000).
629: The peak A$_V$ is located 6$''$ east and 12$''$ south of this position.
630: }
631: \end{figure*}
632: \end{center}
633: \begin{center}
634:
635: \begin{figure*}
636: %\includegraphics[height=7cm]{f2.eps}
637: %\includegraphics[height=21cm]{b68_fig2.eps}
638: %\includegraphics[height=6.5cm]{x.eps}
639: %\plotone{b68_fig2.eps}
640: \plotone{f2.eps}
641: \caption{
642: (a) C$^{18}$O J=1-0 integrated intensity as a function of visual extinction for
643: the entire B68 dark cloud. Inset to the lower right is a similar plot for
644: \CsvO . (b) N$_2$H$^+$ J=1-0 integrated emission as a function
645: of visual extinction in B68. Inset to the lower right are the integrated intensities
646: (in K km s$^{-1}$) of the 2 \NtwoHp\ J=3-2 detections at the corresponding extinctions.
647: In all plots the data are presented as open
648: squares with error bars while solid curves represent the emission predicted by
649: a model combining chemistry with a Monte-Carlo radiative transfer code
650: (\S 3.2). To create these figures the visual
651: extinction data was convolved to a angular resolution of 25$''$ and
652: sampled on the same grid as the molecular observations. For \NtwoHp\ J=3-2 the
653: molecular data and model both have an angular resolution of 10$''$.
654: }
655: \end{figure*}
656: \end{center}
657:
658: \begin{figure*}
659: \includegraphics[height=25cm]{f3.eps}
660: \vspace{-5.0in}
661: %\epsscale{0.80}
662: %\plotone{b68_fig3_2.eps}
663: \caption{
664: Molecular abundances relative to H$_2$ as a
665: function of visual extinction for the best fit model of B68.
666: The abundance of \NtwoHp\ declines by a factor of 2 from A$_V = 2$ mag to
667: A$_V = 17$ mag.
668: }
669: \end{figure*}
670: \end{document}
671:
672:
673:
674:
675:
676: