1: \documentclass[oldversion]{aa}
2: \usepackage{natbib}
3: \usepackage{epsfig}
4: \usepackage{xspace}
5: \usepackage{amssymb}
6: \usepackage{amsmath}
7: \usepackage{graphicx}
8:
9: \def\rhogas{\ensuremath{\rho}\xspace}
10: \def\rhodust{\ensuremath{\rho_\mathrm{d}}\xspace}
11: \def\mugas{\ensuremath{\mu_\mathrm{gas}}\xspace}
12: \def\mp{\ensuremath{m_\mathrm{p}}\xspace}
13: \def\Hp{\ensuremath{H_\mathrm{p}}\xspace}
14: \def\Hs{\ensuremath{H_\mathrm{s}}\xspace}
15: \def\Omegak{\ensuremath{\Omega_\mathrm{K}}\xspace}
16: \def\Mstar{\ensuremath{M_\star}\xspace}
17: \def\Sc{\ensuremath{\mathrm{Sc}}\xspace}
18: \def\St{\ensuremath{\mathrm{St}}\xspace}
19: \def\comma{\,,}
20: \def\fullstop{\,.}
21: %
22: \def\rcentr{R_{\mathrm{centr}}\xspace}
23: \def\rcentrt{R_{\mathrm{centr}}(t)\xspace}
24: \def\rwave{R_{\mathrm{wave}}\xspace}
25: \def\abun{a\xspace}
26: \def\dminf{\dot M_{\mathrm{infall}}\xspace}
27: \def\mcore{M_{\mathrm{core}}\xspace}
28: \def\rcore{R_{\mathrm{core}}\xspace}
29: \def\bcs{{\bar c}_s\xspace}
30: \def\mdisk{M_{\mathrm{disk}}\xspace}
31: \def\mstar{M_{*}\xspace}
32: %
33: \newcommand{\Msun} {M$_\odot$}
34: \newcommand{\Lsun} {L$_\odot$}
35: \newcommand{\macc} {$\dot M_{\mathrm{acc}}$}
36: \newcommand{\Ha} {H$\alpha$}
37: \newcommand{\Roph} {$\rho$~Oph}
38: \newcommand{\simless}{\mathbin{\lower 3pt\hbox
39: {$\rlap{\raise 5pt\hbox{$\char'074$}}\mathchar"7218$}}}
40: \newcommand{\simgreat}{\mathbin{\lower 3pt\hbox
41: {$\rlap{\raise 5pt\hbox{$\char'076$}}\mathchar"7218$}}}
42:
43: \begin{document}
44: \title{Dust crystallinity in protoplanetary disks:
45: the effect of diffusion/viscosity ratio}
46: \titlerunning{Dust crystallinity in protoplanetary disks}
47: \author{Ya.~Pavlyuchenkov \and C.~P. Dullemond}
48: \offprints{Ya.~Pavlyuchenkov, \email{pavyar@mpia.de}}
49: \institute{Max Planck Institut f\"ur Astronomie, K\"onigstuhl 17, 69117
50: Heidelberg, Germany}
51:
52: \abstract{
53: The process of turbulent radial mixing in protoplanetary disks has
54: strong relevance to the analysis of the spatial distribution of
55: crystalline dust species in disks around young stars and to studies of
56: the composition of meteorites and comets in our own solar system.
57: A debate has gone on in the recent literature on the ratio of the
58: effective viscosity coefficient $\nu$ (responsible for accretion)
59: to the turbulent diffusion coefficient $D$ (responsible for mixing).
60: Numerical magneto-hydrodynamic simulations have yielded values between
61: $\nu/D\simeq 10$ (Carballido, Stone \& Pringle \citeyear{carballido:2005})
62: and $\nu/D\simeq 0.85$ (Johansen \& Klahr \citeyear{johansenklahr:2005}).
63: Here we present two analytic arguments for the ratio $\nu/D=1/3$ which
64: are based on elegant, though strongly simplified assumptions. We argue
65: that whichever of these numbers comes closest to reality may be determined
66: {\em observationally} by using spatially resolved mid-infrared
67: measurements of protoplanetary disks around Herbig stars. If meridional
68: flows are present in the disk, then we expect less abundance of
69: crystalline dust in the surface layers, a prediction which can likewise
70: be observationally tested with mid-infrared interferometers.
71: \keywords{accretion disks --- (stars:) formation}}
72: \maketitle
73:
74: \section{Introduction}
75: Turbulent mixing plays an important role in the physics of
76: protoplanetary accretion disks. The same turbulence that is responsible
77: for the anomalous viscosity of the disk (and thus for the accretion
78: process) is also responsible for the radial and vertical mixing of
79: material in the disk. This mixing is likely to have strong influence on
80: thermochemical processes and grain growth in disks. For instance, models
81: of the gas-phase chemistry in disks turn out to be strongly dependent on
82: the level of turbulent mixing (Gammie \citeyear{gammie:1996};
83: Semenov, Wiebe \& Henning \citeyear{semenovwiebe:2004}; Ilgner
84: \citeyear{ilgnernelson:2006a}). This is because some chemical
85: reactions may be very slow in some regions but fast in another. Thus
86: mixing can transport processed gas from the latter region to the former
87: region, affecting the molecular abundances there.
88:
89: For dust solid-state chemistry mixing plays presumably an even bigger role.
90: Some evidence from meteoritics points to weak mixing, such as the recently
91: discovered chemical complementarity of chondrules and matrix in some chondrites
92: (Klerner \& Palme \citeyear{klernerpalme:2000}). On the other hand, the
93: existence of Calcium-Aluminium-rich Inclusions (CAIs) in many chondritic
94: meteorites points toward some level of mixing in the early solar system,
95: since CAIs are thought to form closer to the sun. Samples returned from
96: comet Wild 2 by the STARDUST mission revealed that CAI-like material
97: can be present in comets, which is interpreted as evidence for radial
98: mixing (Zolensky et al.~\citeyear{zolensky:2006}). In infrared
99: observations of protoplanetary disks there is strong evidence of the
100: existence of crystalline silicates at radii in the disk that are far
101: larger than the radius at which the disk temperatures are high enough
102: for thermal annealing. One idea is that they got there by radial mixing
103: from these hot inner regions to the cooler outer regions (Gail
104: \citeyear{gail:2001}; Bockelee-Morvan \citeyear{bockmorvan:2002}).
105:
106: The process of radial mixing by turbulence can be described by a
107: diffusion equation governed by a diffusion coefficient $D$ (Morfill \&
108: V\"olk~\citeyear{morfillvoelk:1984}). Turbulence also plays an important
109: role in the theory of coagulation of dust (e.g.~V\"olk et
110: al.~\citeyear{voelkmorroejon:1980}). It can both prevent and accelerate
111: grain growth, in complex ways. The process of accretion is, on the other
112: hand, described by the equations of mass- and angular momentum
113: conservation in a viscous medium, which is ruled by the effective
114: viscosity coefficient $\nu$ (Shakura \& Sunyaev
115: \citeyear{shaksuny:1973}; Lynden-Bell \& Pringle
116: \citeyear{lyndenpring:1974}). Both $\nu$ and $D$ may depend on distance to
117: the star (i.e.~the radial coordinate of the disk $R$) and both have the
118: same dimension. In fact, because they both arise from the same
119: turbulence, there are reasons to believe that they are nearly the same,
120: apart from a dimensionless factor of order unity which is called the
121: Schmidt number: $\nu/D\equiv\Sc$. The effective viscosity $\nu$
122: may be partly produced by Reynold stress (turbulent motions of the gas),
123: Maxwell-stress (magnetic field lines transporting angular momentum),
124: as well as gravity waves in moderately gravitationally unstable disks.
125: However, the mixing $D$ can {\em only} be due to Reynold stress
126: (turbulent motions). So $D$ and $\nu$ are clearly related, but do not
127: necessarily have to be equal.
128:
129: As has been shown by Clarke \& Pringle~(\citeyear{clarkepringle:1988}), the
130: efficiency of outward radial mixing in steady disks depends on the {\em
131: ratio} of $\nu$ to $D$, i.e.~on $\Sc$, and therefore the Schmidt number
132: plays an essential role in the understanding of the distribution of various
133: chemical and dust species in disks, much more so than the absolute values of
134: $D$ and $\nu$. The Schmidt number of turbulent flows has been discussed in
135: several studies in the past (e.g.~Tennekes \& Lumley
136: \citeyear{tennekeslumley:1972}; McComb \citeyear{mccomb:1990}). These
137: authors typically argue for $\nu/D\simeq 0.7$. It is not clear, however,
138: whether these findings can be directly translated to turbulence in rotating
139: Keplerian accretion disks. Accretion disk flows are fundamentally different
140: from laboratory flow (Balbus \citeyear{balbusreview:2003}).
141: Stone \& Balbus~(\citeyear{StoneBalbus:1996}) have shown, using 3D hydrodynamical
142: simulations as well as analytical arguments, that hydrodynamical turbulence
143: resulting from vertical convection tends to transport angular momentum inward
144: instead of outward. This stands in contrast to turbulence in planar shear
145: flows which, as Stone \& Balbus confirm in their simulations, transports
146: momentum in the opposite direction. The same kind of inward angular momentum
147: transport was reported by R\"udiger et al.~(\citeyear{ruedigeregorov:2005}).
148:
149: Currently the most widely accepted theory for the origin of effective viscosity
150: in accretion disks is that of magneto-rotational turbulence. This picture
151: is based on an interplay between weak magnetic fields and the
152: differentially rotating gas in the disk, causing an instability that
153: drives the turbulence (Balbus \& Hawley~\citeyear{balbushawley:1991}).
154: Angular momentum is then transported both by Reynold stresses as well
155: as by Maxwell stresses. However, there are various uncertainties about whether
156: the magneto-rotational instability can operate in protoplanetary disks
157: which are often very near to being neutral
158: (e.g.~Semenov et al.~\citeyear{semenovwiebe:2004}; Ilgner \& Nelson~\citeyear{ilgnernelson:2006a},
159: Oishi et al.~\citeyear{oishi:2007}). All in all it is still quite unclear
160: what the nature of the turbulence and the mechanism of angular momentum
161: transport is, and therefore the issue of the $\nu/D$ ratio also remains open.
162:
163: Recently, several publications have attempted to shed light on the $\nu/D$
164: ratio in accretion disks with full 3D MHD simulations of such turbulence.
165: These simulations have yielded values ranging from $\nu/D\simeq 10$
166: (Carballido et al.~\citeyear{carballido:2005}) to $\nu/D\simeq
167: 0.85$ (Johansen \& Klahr~\citeyear{johansenklahr:2005}). The drawback of
168: these 3D MHD slab geometry models is that they have a finite numerical
169: resolution which is in general rather coarse. There are reasons to believe
170: that simulations with higher resolution might yield different results.
171: Such initial calculations seem to indeed indicate a reduction of
172: magneto-rotational instability, (Dzyurkevich 2007, private communication).
173:
174: At the same time, new 2D models of Keller \& Gail~(\citeyear{keller:2004})
175: and Tscharnuter \& Gail~(\citeyear{tscharnuter:2007}) show the presence of
176: large-scale circulations within the disk. These calculations support
177: the results found previously in the analytical paper of Urpin~(\citeyear{urpin:1984}),
178: and in the asymptotic study of Regev \& Gitelman~(\citeyear{regev:2002}).
179: Following these results, beyond the certain critical radius near the disk's
180: equatorial plane, the material moves in the outward direction, whereas the
181: accretion flow develops in the surface layer of the disk. As a result of
182: the large-scale circulation, which is driven by viscous angular momentum
183: transfer, advective transport dominates diffusive mixing in the outer part
184: of the disk. Species that are produced or undergo chemical reactions in the
185: warm inner zones of the disks are advectivelly transported into the
186: cool outer regions. Such dynamics cannot be purely represented by any
187: 1D viscous or diffusional models. Thus, $\nu/D$ ratio as direct measure
188: of the mixing efficiency is not appropriate in this picture. But,
189: despite the beauty of this theory, it is based on the postulated
190: anomalous $\alpha$-viscosity, which is an apriory micro-scale quantity.
191: If strong turbulence exists on the scale which is comparable to
192: the height of the disk, it may destroy such a circulation pattern.
193:
194: The question is now: which scenario of the disk evolution (viscous, diffusion
195: or advective) is the right one in nature? From theoretical arguments it is
196: still challenging to tell. But we propose here an observational test that
197: may possibly distinguish between all these cases. This allows us
198: to `measure' effective $\nu/D$ for stars as bright as Herbig Ae stars.
199: Unfortunately this will not work for T Tauri stars and Brown Dwarfs,
200: due to lack of spatial resolution.
201:
202: This paper is organized as follows. In Sect.~\ref{sec-deriv-packages}
203: we re-derive the Pringle (\citeyear{pringle:1981}) equation on the basis of
204: a simple diffusion recipe, and we thereby obtain the Schmidt number under the
205: assumption that the simple diffusion recipe describes the true motion of gas
206: parcels in the disk. In Sect.~\ref{sec-deriv-independent} we derive the
207: Schmidt number in a different way, by assuming that each component of the
208: fluid obeys the same equation. We obtain the same Schmidt number as in
209: Sect.~\ref{sec-deriv-packages}. In Sect.~\ref{sed-abundance-powerlaw} we
210: review how the Schmidt number affects the abundance of crystalline silicates
211: in the inner disk regions. In Sect.~\ref{sec-measuring} we present
212: radiative transfer results for the mid-infrared spectra of spatially
213: resolved disks, and propose how the study of such spectra might provide
214: insight into what is the value of the Schmidt number in real protoplanetary
215: disks.
216:
217: In this paper we focus purely on the diffusion of gas or of small particles
218: that are well coupled to the gas. Bigger dust particles (in the size range
219: of 1 cm or bigger) will decouple from the turbulence and drift inward. This
220: is another process which we do not include in this paper. Also, for
221: simplicity we assume that the disk is vertically averaged and axially
222: symmetric.
223:
224: \section{The diffusion equation for the Keplerian disk:
225: derivation from a discrete model}\label{sec-deriv-packages}
226:
227: In this section we will re-derive the well known equation
228: of Pringle~(\citeyear{pringle:1981}) for the spreading of the accretion disk.
229: The derivation is based on the idea that the evolution of the disk can be
230: completely described by a special kind of turbulent motion of fluid parcels.
231: In normal turbulence each eddy of the disk has a deviation $\delta v$ from the
232: average velocity of the fluid. These are generally non-Keplerian. But the
233: conservation of angular momentum also prevents these gas parcels from drifting
234: too far inward or outward, unless they exchange angular momentum with
235: neighboring gas parcels, which leads to angular momentum transfer in the wrong
236: direction (see above). However, it is known that weakly magnetized disks are
237: unstable to the magneto-rotational instability
238: (Balbus \& Hawley~\citeyear{balbushawley:1991}). Magnetic field
239: lines enable two parcels at different radii to exchange angular momentum
240: `over a distance', if they are thread by the same field lines. We suggest
241: an extremely simplified picture for the motions of these parcels: Consider
242: two parcels, one slowly drifting inward, the other slowly drifting
243: outward. We assume that they both have a locally Keplerian angular velocity
244: at all times. This means that the inward moving parcel loses angular
245: momentum while the other gains angular momentum. We assume that neither
246: parcel has angular momentum exchange with its surroundings. So angular
247: momentum can only be conserved if the loss of angular momentum of the inward
248: moving parcel is compensated by the gain of angular momentum by the outward
249: moving parcel. We invoke an unspecified long-range force that enables this
250: exchange of angular momentum between the two parcels. Physically, the only
251: long-range force that might facilitate this is the Lorenz force, by magnetic
252: fields, so we assume that magnetic fields indeed provide this torque.
253: The accretion and spreading of the disk is, in this picture,
254: nothing other than a material diffusion: Packages of gas being transferred
255: randomly, under the condition of local angular momentum conservation. The
256: radial mixing of a passive tracer in the disk is therefore an integral part
257: of this. If the accretion/spreading of the disk is then viewed in a
258: framework of `viscous disks', then the $\nu/D$ ratio follows strictly from
259: this derivation.
260:
261: This picture was also presented in
262: Tutukov \& Pavlyuchenkov~(\citeyear{tutukovPav:2004}), where different
263: types of astrophysical disks are described in frames of non-stationary
264: diffusional models. However, they used only numerical models and
265: did not provide the equation which governs the evolution of such
266: disks.
267:
268: In this section we will derive these equations and cast them in the
269: standard form of viscous disk equations, which yields the ratio $\nu/D$. We
270: are fully aware that the simplifications of our assumptions are quite
271: drastic and that realistic MHD simulations are presumably better and yield
272: somewhat different results. Nevertheless, we justify our approach in
273: light of the fundamental nature of these equations.
274:
275:
276: \subsection{The mass fluxes from the cell}
277: Let the mass of the disk be small enough so its self-gravitation
278: can be neglected. The disk is assumed to be geometrically thin,
279: Keplerian, and axially symmetric. We describe the structure of the
280: disk in terms of the surface density $\Sigma (R)$, where $R$ is the
281: distance from the axis of symmetry.
282:
283: We introduce a grid \{$R_i$\}, which divides the domain into annular
284: elements which form the grid cells. We assume that the cell size
285: $h_i$=($R_{i+1}$-$R_{i}$) is equal to the {\it characteristic radial mean
286: free path of a turbulent element}. The surface density $\Sigma_i$ is assumed
287: to be constant within each cell. Along with $\Sigma_i$, the average orbital
288: Keplerian velocity is also determined in each grid cell, $V_i$, as well as
289: the average radial component of the turbulent velocity,
290: $V_{r,\mathrm{turb}}(i)$.
291:
292: Let us consider a current cell and denote the nearest cells A and B, see
293: Fig.~\ref{scheme1}.
294: \begin{figure}[b]
295: \centering
296: \includegraphics[width=0.45\textwidth]{./figures/scheme1.eps}
297: \caption{The scheme of the mass transfer in the model
298: of diffusion disk.}
299: \label{scheme1}
300: \vspace{0.5cm}
301: \end{figure}
302: We suppose that in a time $\Delta t$ the mass $M$ in the
303: current cell moves to the neighboring cells due to turbulent motion. The
304: matter leaving this cell is transferred to the two adjacent cells, and this
305: redistribution of the mass must conserve mass and angular momentum according
306: to the philosophy described above. The corresponding equations for the
307: system of three cells can be written
308: \begin{align}
309: &M = M_a+M_b \label{eq1} \\
310: &M R V = M_a R_a V_a + M_b R_b V_b, \label{eq2}
311: \end{align}
312: where $V_a$ and $V_b$ are the Keplerian velocities for radii
313: $R_a$ and $R_b$, respectively.
314:
315: Here we will derive the differential equation which describes the
316: evolution of such a disk in the limit $h \rightarrow 0$, assuming also
317: for simplicity that $h_i$=$h$=const.
318: Using the Taylor expansion of Eq.~\eqref{eq2} we obtain
319: (see Appendix~A1):
320: \begin{align}
321: &M_a = \dfrac{M}{2}\left(1+\dfrac{1}{4}\dfrac{h}{R}\right) \label{eq6} \\
322: &M_b = \dfrac{M}{2}\left(1-\dfrac{1}{4}\dfrac{h}{R}\right). \label{eq7}
323: \end{align}
324: Note that this looks as if mass is always transported outward,
325: since $M_b < M_a$. However, as we shall see below, the true radial
326: flux of matter may have either sign.
327:
328: \subsection{The mass flux across the cell boundary}
329: Denote $M_a^{(R)}$ as a mass transferred from the cell with $R$ into the
330: cell $R+h$. Also, let $M_b^{(R+h)}$ be a mass transferred from the cell
331: $R+h$ into the cell $R$, see Fig.~\ref{scheme2}.
332: \begin{figure}
333: \centering
334: \includegraphics[width=0.3\textwidth]{./figures/scheme2.eps}
335: \caption{The scheme of the mass transfer between two cells
336: in the model of diffusion disk}
337: \label{scheme2}
338: \vspace{0.5cm}
339: \end{figure}
340: Denote also $F_{ab}$ as a total flux across the boundary
341: between the cells:
342: \begin{equation}
343: F_{ab}\Delta t=M_a^{(R)}-M_b^{(R+h)}. \label{eq10}
344: \end{equation}
345: Using the Taylor expansion of Eq.~\eqref{eq10} (see Appendix~A2) we find:
346: \begin{equation}
347: F_{ab} \Delta t = - \dfrac{1}{2}h R^{1/2} \dfrac{\partial}{\partial R}(MR^{-1/2}).
348: \label{eq12}
349: \end{equation}
350: Note, that if the mass flux $M^{(R)}$ does not depend on radius, then
351: $F_{ab}>0$, i.e. the mass is transferred outwards.
352:
353: One of the basic assumptions of our diffusion model is the relation
354: for the flux from the cell:
355: \begin{equation}
356: M = 2\pi R \cdot \Sigma \cdot V_{r,\mathrm{turb}} \cdot \Delta t.
357: \label{eq13}
358: \end{equation}
359: If we substitute this relation into Eq.~\eqref{eq12} and introduce
360: the diffusion coefficient (see Appendix~A3 for the choice of numerical
361: coefficient):
362: \begin{equation}
363: D= \frac{1}{2} h \cdot V_{r,\mathrm{turb}},
364: \label{eq14}
365: \end{equation}
366: we get the final formula for the total flux across the cell boundary:
367: \begin{equation}
368: F_{ab} = -\pi R^{1/2} \dfrac{\partial}{\partial R}(D\Sigma R^{1/2}).
369: \label{eq15}
370: \end{equation}
371: Note now that if $\Sigma$ and $D$ do not depend on radius than $F_{ab}=-\pi
372: D \Sigma <0$ i.e. the disk becomes an {\em accretion} disk.
373:
374: \subsection{The equation of the diffusion disk evolution}
375: The law of the mass conservation means:
376: \begin{equation}
377: 2\pi R \Delta \Sigma = - \dfrac{\partial F_{ab}}{\partial R}\Delta t,
378: \label{eq17}
379: \end{equation}
380: Using Eq.~\eqref{eq15} we find finally:
381: \begin{equation}\label{eq-lbp-diff}
382: \frac{\partial\Sigma}{\partial t} = \frac{1}{R}\frac{\partial }{\partial R}
383: \left[\sqrt{R}\frac{\partial }{\partial R}\left(\Sigma D \sqrt{R}\right)\right].
384: \end{equation}
385:
386: This is the required equation which describes the evolution of the diffusive
387: disk.
388:
389: If we take $\nu=D/3$ then we can cast Eq.~(\ref{eq-lbp-diff}) into the
390: familiar equation for accretion disk evolution (see Pringle
391: \citeyear{pringle:1981}):
392: \begin{equation}\label{eq-fullsigma-evol}
393: \frac{\partial\Sigma}{\partial t} = \frac{3}{R}\frac{\partial }{\partial R}
394: \left[\sqrt{R}\frac{\partial }{\partial R}\left(\Sigma \nu \sqrt{R}\right)\right]
395: \end{equation}
396: This equation can be derived from the Navier-Stokes equations for viscous
397: disks, where $\nu$ is the standard viscosity. Since $D$ is the true
398: diffusion coefficient of the matter, we have found that the ratio $\nu/D$
399: must be 1/3 in our picture.
400:
401:
402: \section{The diffusion equation for Keplerian disk: multicomponent
403: medium}\label{sec-deriv-independent}
404: Another way to `derive' the ratio of $\nu/D$ is to start from the {\em
405: Ansatz} that each gas or dust species evolves independently according to:
406: \begin{equation}\label{eq-each-species-indep}
407: \frac{\partial\Sigma_i}{\partial t} = \frac{3}{R}\frac{\partial }{\partial
408: R} \left[\sqrt{R}\frac{\partial }{\partial R}\left(\Sigma_i \nu
409: \sqrt{R}\right)\right]
410: \end{equation}
411: where $\Sigma_i$ is the surface density of species $i$. The $\Sigma_i$
412: obey completeness:
413: \begin{equation}
414: \Sigma_1 + \Sigma_2 + \cdots + \Sigma_n = \Sigma
415: \end{equation}
416: where $n$ is the total number of species.
417:
418: The underlying argument to support the Ansatz of
419: Eq.~(\ref{eq-each-species-indep}) is the following: Suppose that gas+dust
420: parcels indeed split up and move inward/outward, as assumed in
421: Sect.~\ref{sec-deriv-packages}. The relative abundances of the various species are
422: the same in the inward moving parcel as in the outward moving parcel (they
423: were both from the same original parcel). This means that if the angular
424: momentum stays constant between the inward and outward moving parcel, the
425: same can be said of the individual species. So: species $i$ in the inward
426: moving parcel loses as much angular momentum as the same species in the
427: outward moving parcel gains. Therefore the diffusion happens in each species
428: identically, and therefore we argue that Eq.~(\ref{eq-each-species-indep})
429: holds for each species separately, independent of the other species. This is
430: even true if one of the species represents 99.9\% of the mass and the other
431: only $0.1\%$.
432:
433: So, starting from Eq.~(\ref{eq-each-species-indep}) we can rewrite:
434: \begin{equation}\label{eq-separ-step1}
435: \begin{split}
436: \frac{\partial\Sigma_i}{\partial t} &= \frac{3}{R}\frac{\partial }{\partial
437: R} \left[\sqrt{R}\frac{\partial }{\partial R}\left(\frac{\Sigma_i}{\Sigma}
438: \Sigma \nu \sqrt{R}\right)\right]\\
439: &=\frac{3}{R}\frac{\partial }{\partial
440: R} \left[\sqrt{R}\frac{\Sigma_i}{\Sigma}\frac{\partial }{\partial R}\left(
441: \Sigma \nu \sqrt{R}\right)
442: +R\Sigma \nu \frac{\partial }{\partial R}\left(
443: \frac{\Sigma_i}{\Sigma}\right)
444: \right]
445: \end{split}
446: \end{equation}
447: If we write Eq.~(\ref{eq-fullsigma-evol}) as a conservation equation as
448: $\dot \Sigma + (1/R)\partial [R\Sigma v_R]/\partial R=0$ where $v_R$ is the
449: average radial velocity of the matter (including all species), then by
450: Eq.~(\ref{eq-fullsigma-evol}) this
451: velocity is given as:
452: \begin{equation}\label{eq-velo}
453: v_R = -\frac{3}{\Sigma\sqrt{R}}\frac{\partial}{\partial R}
454: \left(\Sigma\nu\sqrt{R}\right)
455: \end{equation}
456: Using Eq.~(\ref{eq-velo}) in Eq.~(\ref{eq-separ-step1}) one obtains:
457: \begin{equation}\label{eq-form-morfill}
458: \frac{\partial\Sigma_i}{\partial t} + \frac{1}{R}\frac{\partial}{\partial R}
459: \left(R\Sigma_i v_R\right) = \frac{3}{R}\frac{\partial}{\partial R}
460: \left[R\nu \Sigma\frac{\partial}{\partial R}
461: \left(\frac{\Sigma_i}{\Sigma}\right)\right]
462: \end{equation}
463: This is the same as the equation introduced by Morfill \& V\"olk~(\citeyear{morfillvoelk:1984})
464: but now with:
465: \begin{equation}
466: \frac{\nu}{D} = 1/3
467: \end{equation}
468:
469: So we have shown here that, provided $\nu/D$=1/3, the picture proposed by
470: Morfill \& V\"olk~(\citeyear{morfillvoelk:1984}) that diffusion takes place
471: in the comoving frame of an otherwise laminar accretion disk is equivalent
472: to our picture of exchanging Keplerian parcels.
473:
474: Again we stress that our derivation is based on a simplifying assumption.
475: One can consider our value of $\nu/D=3$ as a minimum value of $\nu/D$.
476: At the other extreme is the picture of a perfectly laminar but
477: viscous disk with $\nu/D=\inf$. The true value must be within these
478: limits.
479:
480: \section{Effect of $\nu/D$ on abundance of crystalline silicates in disks}
481: \label{sed-abundance-powerlaw}
482: Although our simple analytic argumentation predicts that $\Sc=1/3$,
483: MHD simulations and experiments with rotating turbulent flows
484: show other ratios.
485: At the moment we need to consider values ranging from $\Sc=1/3$ (this study)
486: to $\Sc=10$ (Carballido, Stone \& Pringle~\citeyear{carballido:2005}; Johansen,
487: Klahr \& Mee~\citeyear{johansenklahrmee:2006}). So how do these various
488: values affect the distribution of molecular species and dust species in
489: the disk? What are the consequences of this, for instance,
490: for the problem of crystalline silicates in disks? One may argue that the
491: difference between $\nu/D$=1/3 and $\nu/D=1$ is `only' a factor of 3 and
492: therefore not very strong: the time scales of diffusion will be
493: different by only a factor of 3. However, for accretion disks in semi-steady
494: state, a difference of 3 in the diffusion constant can have very strong
495: effects on the distribution of thermally processed (crystalline) particles
496: in the disk. If we assume that particles are only thermally processed in the
497: very inner regions of the disk (i.e.~close to the star) then to find
498: particles at larger radii, they have to diffuse outward (e.g.~Gail
499: \citeyear{gail:2001}; Bockelee-Morvan \citeyear{bockmorvan:2002}). The
500: efficiency of this outward diffusion is very dependent on the $\nu/D$ ratio.
501: This is due to the fact that the trace particles (crystalline dust) have to `swim
502: upstream' against the accretion flow to reach larger radii. This was modeled
503: by Clarke \& Pringle~(\citeyear{clarkepringle:1988}) who showed that the
504: abundance of such a tracer as a function of radius can be derived
505: analytically for steady accretion disks. Since in the absence of disk
506: instabilities the inner regions of viscously evolving disks change over a
507: time scale much larger than the local viscous time, these inner regions
508: (say, inward of a few tens of AU at an age of 1 Myr) can be regarded as
509: being in semi-steady state. This simplifies the analysis drastically. We can
510: then start from the standard advection-diffusion equation, after Morfill \&
511: V\"olk~(\citeyear{morfillvoelk:1984}):
512: \begin{equation}\label{eq-morfill}
513: \frac{\partial\Sigma_i}{\partial t} + \frac{1}{R}\frac{\partial}{\partial R}
514: \left(R\Sigma_i v_R\right) = \frac{1}{R}\frac{\partial}{\partial R}
515: \left[RD\Sigma\frac{\partial}{\partial R}
516: \left(\frac{\Sigma_i}{\Sigma}\right)\right]
517: \end{equation}
518: For the special case of $\nu/D$=1/3 this becomes equal to
519: Eq.~(\ref{eq-form-morfill}), but we will retain the more general form of
520: Eq.~(\ref{eq-morfill}) here. For a steady state disk we have (for $R\gg
521: R_{\mathrm{in}}$ where $R_{\mathrm{in}}$ is the inner radius of the disk):
522: \begin{equation}\label{eq-steadystate1}
523: \dot M = - 2\pi R\Sigma v_R,
524: \end{equation}
525: with
526: \begin{equation}\label{eq-steadystate2}
527: v_R = -\frac{3}{2}\frac{\nu}{R}.
528: \end{equation}
529: Here $\dot M$ is the accretion rate of the disk. With the Schmidt number
530: $\Sc\equiv \nu/D$, by using Eqs.~(\ref{eq-steadystate1}-\ref{eq-steadystate2}),
531: and by taking $\partial\Sigma_i/\partial t=0$ (steady state) we can rewrite
532: Eq.~(\ref{eq-morfill}) in the form:
533: \begin{equation}
534: \frac{\partial}{\partial\lg R}\left(\frac{\Sigma_i}{\Sigma}\right)
535: =-\frac{2}{3}\frac{1}{\Sc}\frac{\partial^2}{\partial(\lg
536: R)^2}\left(\frac{\Sigma_i}{\Sigma}\right)
537: \end{equation}
538: The solution is:
539: \begin{equation}\label{eq-mix-sol1}
540: \frac{\Sigma_i}{\Sigma} = \sigma_i^{(0)}
541: + \sigma_i^{(1)} \left(\frac{R_{1}}{R}\right)^{\frac{3}{2}\Sc}
542: \end{equation}
543: This is a general steady-state solution for mixing of any tracer in the disk
544: for $R\gg R_{\mathrm{in}}$. The $\sigma_i^{(0)}$ is a background abundance while
545: $\sigma_i^{(1)}$ is the abundance at some radius $R_1$.
546:
547: In general, the Eq.~\eqref{eq-mix-sol1} is appropriate only if no
548: large-scale radial flows are present in the disk. However, it is
549: possible to modify this equation in a way that it approximately
550: accounts for the effect of large-scale advective flows onto emergent spectra.
551: Following the results of Keller \& Gail~(\citeyear{keller:2004}) and
552: Tscharnuter \& Gail~(\citeyear{tscharnuter:2007})
553: we assume that near the disk's midplane the material moves outward,
554: while in the surface layers the material flows inward.
555: Since we assume that the surface layer of the disk is responsible for
556: the emergent spectra we change the Eq.~\eqref{eq-mix-sol1}
557: so as it describes the abundances of the tracers in the surface
558: layer. Therefore, we substitute $V_R(1+C)$ for $V_R$ in the
559: Eq.~\eqref{eq-steadystate1}, where $C$ is the circulation velocity in
560: units of the vertically averaged accretion velocity, and thus we
561: come to a more general relation for the tracer abundances in the surface layer:
562: \begin{equation}\label{eq-mix-sol2}
563: \frac{\Sigma_i}{\Sigma} = \sigma_i^{(0)}
564: + \sigma_i^{(1)} \left(\frac{R_{1}}{R}\right)^{\frac{3}{2}\Sc\,(1+C)}
565: \end{equation}
566: We suppose that $C$ can easily be 2-3 or more which has major
567: consequences for the mixing. Given this modification, we introduce
568: the effective Schmidt number:
569: \begin{equation}
570: \bar{\Sc}=\Sc(1+C).
571: \end{equation}
572: Note, that large values of $C$ `mimic'\, large Schmidt numbers, $\Sc$,
573: in the picture with no circulation.
574:
575: Let us now apply these results to the problem of crystallinity of dust in a
576: disk which is hot enough for thermal annealing inward of
577: $R_{\mathrm{anneal}}$. Since the dependence of the efficiency of thermal
578: annealing on temperature is extremely sharp, one can say that inward of
579: $R_{\mathrm{anneal}}$ all dust is crystalline. For $R>R_{\mathrm{anneal}}$
580: there is no annealing and the mixing solution Eq.~(\ref{eq-mix-sol2}) is
581: valid in which $R_1\equiv R_{\mathrm{anneal}}$ and
582: $\sigma_{\mathrm{cryst}}^{(1)}=1$. The background crystallinity
583: $\sigma_{\mathrm{cryst}}^{(0)}$ is the globally present level of
584: crystallinity, discussed below. One sees that the abundance of crystalline
585: silicates at any given radius depends only on $R_{\mathrm{anneal}}$, the
586: base level of crystallinity $\sigma_{\mathrm{cryst}}^{(0)}$ and the value of
587: {$\bar{\Sc}$}. It does {\em not} depend on the value of $\nu$ itself
588: (i.e.~not on the value of $\alpha$). For strong or weak turbulence the
589: results are the same, as long as $\bar{\Sc}$ remains the same.
590: The solutions to Eq.~(\ref{eq-mix-sol2})
591: for $\sigma_{\mathrm{cryst}}^{(0)}=0$ (no global crystallinity) and
592: for $\sigma_{\mathrm{cryst}}^{(0)}=0.1$ (10\% global crystallinity)
593: are shown in Fig.~\ref{fig-abundance}.
594: \begin{figure}
595: \centering
596: \includegraphics[width=0.45\textwidth]{./figures/crystalinity_log.eps}
597: \caption{\label{fig-abundance}The abundance of crystalline silicates in a
598: steadily accreting protoplanetary disk around a Herbig star of $M_{*}=2.5
599: M_{\odot}$, $R_{*}=2.5 L_{\odot}$ and $T_{*}=10^4$K. Solid lines: assuming
600: that {\em only} thermal annealing in the warm inner regions (here: inward of
601: about $1$ AU) can produce crystalline silicates. Dotted lines: assuming that
602: some other mechanism has produced a global base level of crystallinity of
603: 10\%.}
604: \end{figure}
605:
606: It is important to keep in mind that these analytic solutions for the level
607: of crystallinity are only valid for $R_1\equiv R_{\mathrm{anneal}}\gg
608: R_{\mathrm{in}}$. This is the case for Herbig stars, since they are so
609: bright that the annealing radius is around 1 AU while the inner disk radius
610: is more than hundred times smaller. For T Tauri stars, on the
611: other hand, the annealing radius is only about ten times larger than the
612: inner radius, which means that the effects of the no-friction boundary
613: condition of the disk at $R_{\mathrm{in}}$ are still strongly affecting the
614: solution, and hence the above analytic solution is not completely correct.
615: Nonetheless, the rough form remains correct even in that case.
616:
617: Now let us focus on the case without a base level of crystallinity. It
618: can be seen that the abundance of crystals at large radii is extremely
619: sensitive to the power index $\bar{\Sc}$. For $\bar{\Sc}=10$, i.e.
620: for large Schmidt numbers or large surface inflow velocity, there is
621: essentially no outward mixing. For $\bar{\Sc}=1$ or $\bar{\Sc}=0.7$
622: the crystals are mixed out to a level of 3\% and 10\% respectively at
623: 10 AU. For $\bar{\Sc}=1/3$ the 10\% crystallinity is reached at 100 AU, i.e.~out
624: to a hundred times larger radius. This shows that the seemingly small
625: differences between the various theoretically derived values of $\bar{\Sc}$ make
626: a very large difference in the distribution of crystalline silicates in
627: the disk.
628:
629: So far we have assumed that crystalline silicates are only produced by
630: thermal annealing in the inner disk regions ($R\le R_{\mathrm{anneal}}$).
631: However, there is an on going debate about what causes dust grains to become
632: crystalline. While the thermal annealing and radial mixing of dust (Gail
633: \citeyear{gail:2001}) is a natural phenomenon that is likely to happen in
634: all disks, there may be additional mechanisms of crystallizing dust such as
635: shock heating (Harker \& Desch~\citeyear{harkerdesch:2002}) or lightning
636: (Gibbard~\citeyear{gibbardlevymorfill:1997} and references
637: therein). Evidence for these alternative mechanisms is found in infrared
638: spectroscopy of protoplanetary disks, where it is found in many sources that
639: the crystalline grains are dominated by forsterite in the outer regions of
640: many disks. This is contrary to the radial mixing models which predict
641: that enstatite dominates in the outer disks (Bouwman et al.~in prep.).
642:
643: From the solar system there is also some evidence that other mechanisms
644: might crystallize dust or dust aggregates (e.g.~review by Trieloff \&
645: Palme~\citeyear{trieloffpalme:2006}). However, the inner regions of disks,
646: near the dust sublimation radius, are always warm enough to assure that
647: thermal annealing takes place. Therefore, independent of the existence of
648: other crystallization processes, we expect with near certainty that inward
649: of some radius all the dust is 100\% crystalline
650: (i.e.~$\sigma_{\mathrm{cryst}}^{(0)}+\sigma_{\mathrm{cryst}}^{(1)}=1$). The
651: other crystallization processes, assuming that they can crystallize dust
652: throughout the disk, produce a base level $\sigma_{\mathrm{cryst}}^{(0)}$ of
653: crystallinity. The dotted lines in Fig.~\ref{fig-abundance} show how the
654: combined effects of crystallization work out. We assume here that the time
655: scales of these additional crystallizing processes are longer than the local
656: viscous time in the region we study here, so that the level of crystallinity
657: in the outer disk regions has been built up over a long time and remain
658: nearly unchanged on the shorter time scales in this inner region.
659:
660: Another source of a `base-level' of crystallinity
661: $\sigma_{\mathrm{cryst}}^{(0)}$ is that thermal annealing may also happen
662: throughout the disk in the very early disk-formation phases (Dullemond, Apai
663: \& Walch~\citeyear{dullapaiwalch:2006}). They showed that in the stage that
664: the disk is formed, depending on initial conditions, much of the outward
665: transport of material may not be due to the `upstream' diffusion, but due to
666: the fact that the disk material itself is spreading outward (Lynden-Bell \&
667: Pringle~\citeyear{lyndenpring:1974}). Later, this pre-processed material
668: accretes back inward and forms a base level of crystallinity
669: $\sigma_{\mathrm{cryst}}^{(0)}$
670: (see Dullemond et al.~\citeyear{dullapaiwalch:2006}, Fig.~2).
671:
672: If large-scale circulations are present in the disk then over a
673: large distance there may be vertical mixing between the outward moving
674: midplane and inward moving surface layers. This is likely to
675: produce an enhanced `base'\, value of crystallinity as it inserts
676: crystalline grains into the surface layers at large radii.
677:
678: The conclusion of this section is that radial mixing can have a strong
679: effect on the distribution of crystalline dust, which is important for
680: understanding infrared observations of protoplanetary disks, as well as
681: for understanding the constituents of meteorites. Radial mixing may also
682: strongly affect disk chemistry or grain growth. A small difference in
683: the assumed value of $\Sc$ as well as of $C$ could make a strong
684: difference in the results of the models.
685:
686: \section{`Measuring'
687: $\bar{\Sc}$ from disks around Herbig Ae stars} \label{sec-measuring}
688: It is clear that it is of high importance to understand which is the precise
689: value of $\bar{\Sc}$. Theories will likely always have
690: uncertainties. In this section we propose a way to derive the value of
691: $\bar{\Sc}$ {\em observationally}. From Fig.~\ref{fig-abundance} it is clear that for
692: Herbig Ae stars the crystallinity of the dust in the region roughly between
693: $1$ AU and $10$ AU is strongly affected by the ratio $\bar{\Sc}$, even if there
694: is a base level $\sigma_{\mathrm{cryst}}^{(0)}$ of crystallinity due to
695: other processes. And in the case of Schmidt numbers below 1 the radial
696: mixing may even affect regions as far as out 100 AU. If we can
697: observationally measure the crystallinity {\em as a function of radius} then
698: we may be able to determine both $\bar{\Sc}$ as well as
699: $\sigma_{\mathrm{cryst}}^{(0)}$.
700:
701: \begin{figure*}
702: \centering
703: \includegraphics[width=0.8\textwidth]{./figures/spec_base_1.ps}
704: \caption{Intensity spectra of the disk (arbitrarily normalized) at different
705: radii, for different assumed $\bar{\Sc}$, for the standard model shown
706: in Fig.~\ref{fig-abundance}. Here the base level of crystallinity
707: $\sigma_{\mathrm{cryst}}^{(0)}$ is chosen to be 0, i.e.~the spectra
708: correspond to the crystallinity levels shown as solid lines in
709: Fig.~\ref{fig-abundance}.}
710: \label{fig-intspec-nobase}
711: \vspace{0.5cm}
712: \end{figure*}
713:
714: \begin{figure*}
715: \centering
716: \includegraphics[width=0.8\textwidth]{./figures/spec_base_2.ps}
717: \caption{Same as Fig.~\ref{fig-intspec-nobase}, but with a base
718: level of crystallinity of $\sigma_{\mathrm{cryst}}^{(0)}=0.1$.}
719: \label{fig-intspec-base}
720: \vspace{0.5cm}
721: \end{figure*}
722:
723: The abundance of crystalline dust in a disk can be determined from the
724: analysis of infrared spectra (see review by Natta et
725: al.~\citeyear{nattappv:2007}). For ground-based observations the 8-13
726: $\mu$m window is well-suited for this. There are by now a number of
727: instruments and telescopes that can spectrally and spatially resolve disks
728: in the 10 micron regime at spatial resolutions of about 20 AU for typical
729: sources: e.g.~VISIR at the VLT, COMICS at Subaru and T-ReCS at
730: Gemini-South. A spatial resolution of about 1 AU can be obtained with
731: mid-infrared interferometry with the MIDI instrument on the VLT (Leinert et
732: al.~\citeyear{leinertvanboekel:2004}). Although in particular the
733: interferometry observations do not directly yield spectra as a function of
734: radius, they do give information on those spatial scales that can be
735: compared to model predictions. Of course these model predictions have to be
736: done for each source individually.
737:
738: To demonstrate the principle, we show
739: the intensity-spectrum at various radii in a model of a disk around a
740: Herbig Ae/Be star of $M_{*}=2.5 M_{\odot}$, $R_{*}=2.5 L_{\odot}$ and
741: $T_{*}=10^4$K, i.e. the same system as was shown in
742: Fig.~\ref{fig-abundance}. We took a disk with a mass of 0.0015 $M_{\odot}$,
743: a size of 90 AU and a surface density profile of $\Sigma(R) \propto R^{-1}$.
744: The 2-D axisymmetric density structure of the model was taken to
745: be $\rho(r,z)=\Sigma(R)\exp(-z^2/2R^2)/(H_p(R)\sqrt{2\pi})$, where $H_p(R)$
746: is the pressure scale height estimated from the midplane temperature of
747: a simple flaring disk model of the kind of Chiang \& Goldreich~(\citeyear{cg97}).
748: We included two dust species, both consisting of 0.1 $\mu$m size grains. The
749: first species is a mixture of 25\% amorphous olivine, 25\% amorphous
750: pyroxene and 50\% carbon. The second species consists of 25\% crystalline
751: enstatite and 25\% crystalline forsterite and 50\% carbon. Within each
752: species the various components are in thermal contact, but the two species
753: are not in thermal contact with each other. We used optical constants from
754: Dorschner et al.~(\citeyear{dorschner:1995}) for the amorphous olivine and
755: pyroxene, from Servoin \& Piriou~(\citeyear{servoin:1973}) for crystalline
756: forsterite and from J\"ager et al.~(\citeyear{jaeger:1998}) for crystalline
757: enstatite. The method of computation of the opacities from these constants
758: is described in Min et al.~(\citeyear{minhove:2005}).
759: Given the density structure, the opacities and the stellar parameters we
760: can apply a Monte-Carlo radiative transfer code
761: (RADMC, see Dullemond \& Dominik~\citeyear{duldomdisk:2004}) to compute the
762: dust temperature everywhere in the disk. RADMC is a Monte Carlo code for
763: dust continuum radiative transfer based on the method of
764: Bjorkman \& Wood~(\citeyear{BW:2001}) combined with the method of
765: Lucy~(\citeyear{Lucy:1999}). This gives the typical structure
766: of irradiated disks which have a warm surface and a cooler interior.
767: A ray-tracing program is then applied to compute, from these density and
768: temperature structures, a spectrum. Due to the warm surface layer lying on
769: top of a cooler interior, the dust features appear in emission in the
770: spectrum.
771: Figure~\ref{fig-intspec-nobase} shows the intensity spectra
772: ($I_\nu(R)$) obtained in this way at
773: various radii (arbitrarily normalized) for four different values of $\Sc$,
774: assuming that there is no base level of crystallinity, i.e.\, that thermal
775: annealing in the inner disk regions is the only source of crystallinity.
776: The intensity spectrum is the intensity of the face-on disk image as
777: seen at that particular radius, as a function of wavelength.
778: In Fig.~\ref{fig-intspec-base} the same is shown, but including a base level
779: $\sigma_{\mathrm{cryst}}^{(0)}$ of crystallinity due to either the disk's
780: formation history (e.g.~Dullemond et al.~\citeyear{dullapaiwalch:2006})
781: or due to additional crystallization processes (e.g.~Harker \& Desch
782: \citeyear{harkerdesch:2002}).
783:
784: From Figs.~\ref{fig-intspec-nobase} and \ref{fig-intspec-base} one can see
785: that both the base level $\sigma_{\mathrm{cryst}}^{(0)}$ and the
786: $\bar{\Sc}$ affect the spectra in a way that would allow the
787: reconstruction of these two values from the observations. The
788: $\sigma_{\mathrm{cryst}}^{(0)}$ is reflected in the fact that the strength
789: of the crystalline features decreases with radius but gets `stuck' at some
790: level which is set by the value of $\sigma_{\mathrm{cryst}}^{(0)}$. The
791: $\bar{\Sc}$ value is reflected in the rapidity by which the crystalline features
792: become weaker with radius as one goes to larger radii. And
793: $R_{\mathrm{anneal}}$ is found as the radius by which the crystallinity
794: suddenly starts to drop. In principle the $R_{\mathrm{anneal}}$ should
795: follow directly from theoretical considerations, as the dust temperature in
796: the disk's surface layer, for a given grain size, can be determined
797: theoretically. However, the grain size is not known beforehand, and an
798: uncertainty of the grain size in the $0.1$ to $1$ $\mu$m regime can not be
799: constrained by the analysis of the $10$ $\mu$m feature shape, but it does
800: influence the dust temperature. Therefore an observational
801: determination of $R_{\mathrm{anneal}}$ is necessary and possible.
802:
803: However, there are some difficulties with this scheme. One obvious problem
804: is that one requires rather accurate measurements of the spectrum at many
805: radii. Since most of the action takes place very close to 1 AU, these
806: measurements would rely strongly on the MIDI interferometer at the VLT or its
807: planned successor, and it would require measurements with at least 4
808: baselines that are all ideally aligned in the same position angle on the
809: sky. Moreover it would be best if the disk is nearly face-on and bright.
810:
811: Another complication is that thermal annealing may take place in the surface
812: layers at a different radius than the midplane. If the disk's heating is
813: dominated by irradiation, then the midplane temperature of the disk is much
814: smaller than the surface temperature. In this case there are radii where the
815: dust is above annealing temperature in the surface, but below annealing
816: temperature in the disk's midplane. It then depends strongly on the {\em
817: vertical} mixing efficiency whether the annealing in the surface can also
818: strongly affect the crystallinity of dust in the midplane. If not, then the
819: $R_{\mathrm{anneal}}$ relevant for the radial mixing might be smaller than
820: the radius where the surface is annealed. One would then expect, as a
821: function of radius, the crystallinity to be unity out to that surface
822: annealing radius and then to suddenly drop to the level of the power-law
823: which originated at a smaller annealing radius. In principle this could also
824: be observed, if sufficiently finely-spaced measurements are taken. If
825: accretional (viscous) heating could heat the disk's midplane sufficiently
826: strongly, then the oppose may be true: the annealing radius would then be
827: larger than the annealing radius computed with irradiation only, and only
828: drop at larger radii. Moreover, if meridional flows are present, the
829: picture might blur even more.
830:
831:
832: It is clear that the full interpretation of these measurements will not be
833: easy. But since the Schmidt number is such a fundamental number for
834: accretion disk theory, it might be worthwhile to attempt to do such
835: measurements nonetheless.
836:
837:
838: \section{Conclusions}
839:
840: In this paper we discuss the ratio between turbulent diffusion
841: and viscosity in protoplanetary disks and study an
842: effect of this ratio on the dust crystallinity. Our conclusions can be
843: summarized as follows:
844:
845: \begin{enumerate}
846:
847: \item The efficiency of radial mixing of any tracers in a steady
848: state disk depends on the ratio between viscosity and diffusion
849: coefficients i.e. on the Schmidt number $\Sc=\nu/D$ if we assume that
850: the disk is well mixed in vertical direction. Therefore, the Schmidt
851: number plays an essential role in the understanding of the distribution
852: of the various chemical and dust species in disks.
853:
854: \item We present the simple analytic model of the diffusively spreading
855: disk. We show that evolution of the diffusively spreading disk
856: and evolution of the viscousely spreading disk are described by the
857: same equations if we assume $\nu/D=1/3$ where $\nu$ is the effective
858: viscosity coefficient in the classical Shakura \& Sunyaev~(\citeyear{shaksuny:1973})
859: picture. This does not mean, however, that we necessarily argue
860: that true accretion disks have this $\Sc=1/3$: this depends on the validity
861: of our assumptions (vertically averaged disk, good mixing between
862: dust and gas, a particular kind of turbulence). However, one can regard the
863: value $\Sc=1/3$ as the lower limit of the Schmidt number which can be included
864: in the models of radial mixing in Keplerian disks.
865:
866: \item If large-scale meridional flows are present in the disks then the
867: Schmidt number cannot be used as a direct measure of the mixing efficiency.
868: However, it is possible to approximately describe the abundance of the
869: tracers in the surface layers of the disk by introducing the
870: effective Schmidt number $\bar{\Sc}=\Sc(1+C)$ where $C$ is the
871: layer velocity in units of vertically averaged accretion velocity. This
872: approximation is valid as long as no vertical mixing between
873: in- and outward moving layers is assumed. In this case, we expect less
874: abundance of crystalline dust in the surface layers.
875:
876: \item We show that the value of $\bar{\Sc}$ may possibly be measured
877: in protoplanetary disks around young stars by using mid-infrared
878: interferometry. This requires measurements of a single source at many
879: baselines which all lie along the same position angle across the sky.
880:
881: \end{enumerate}
882:
883: \begin{acknowledgements}
884: We are grateful to E.~Kurbatov and A.V.~Tutukov for help in derivation
885: of the diffusion equation. We are grateful to N.~Dzyurkevich for
886: a careful review of the paper. We also wish to thank R.~van Boekel,
887: J.~Bouwman and M.~Min for useful discussions and for the opacity
888: tables. We thank the anonymous referee for very insightful comments
889: which have significantly improved the paper and reminded us of the
890: importance of circulation in disks.
891: \end{acknowledgements}
892:
893: \section*{Appendix A1}
894: Consider Eq.~\eqref{eq2}, and produce the Taylor expansion around central cell, $R$:
895: \begin{eqnarray}
896: M R V = M_a (R+h)\left(V+\dfrac{\partial V}{\partial R} h + \dfrac{1}{2} \dfrac{\partial^{2}V}{\partial R^{2}} h^{2}\right) \nonumber \\
897: + M_b (R-h)\left(V-\dfrac{\partial V}{\partial R} h + \dfrac{1}{2} \dfrac{\partial^{2}V}{\partial R^{2}} h^{2}\right). \label{eq3a}
898: \end{eqnarray}
899: After elementary operations, using Eq.~\eqref{eq1} and dropping terms
900: of order $h^3$ we get:
901: \begin{equation}
902: (M_b-M_a) = \dfrac{Mh}{2}\frac{\dfrac{\partial^2}{\partial R^2}\left(RV\right)}
903: {\dfrac{\partial}{\partial R}\left(RV\right)}. \label{eq4a}
904: \end{equation}
905: For the Keplerian disk, $V \sim R^{-1/2}$, this relation becomes:
906: \begin{equation}
907: (M_b-M_a) = -\dfrac{M}{4}\dfrac{h}{R}. \label{eq5a}
908: \end{equation}
909: Using Eq.~\eqref{eq1} we obtain:
910: \begin{align}
911: &M_a = \dfrac{M}{2}\left(1+\dfrac{1}{4}\dfrac{h}{R}\right) \label{eq6a} \\
912: &M_b = \dfrac{M}{2}\left(1-\dfrac{1}{4}\dfrac{h}{R}\right). \label{eq7a}
913: \end{align}
914:
915: \section*{Appendix A2}
916: Rewrite the fluxes $M_a^{(R)}$ and $M_b^{(R+h)}$ using Eq.~\eqref{eq6} and~\eqref{eq7}:
917: \begin{align}
918: &M_a^{(R)} = \dfrac{1}{2}M^{(R)}\left\{1+\dfrac{1}{4}\dfrac{h}{R}\right\} \label{eq8a} \\
919: &M_b^{(R+h)} = \dfrac{1}{2}M^{(R+h)}\left\{1-\dfrac{1}{4}\dfrac{h}{(R+h)}\right\}. \label{eq9a}
920: \end{align}
921: where $M^{(R)}$ and $M^{(R+h)}$ are the masses emerging from the cells
922: with radii $R$ and $R+h$ respectively.
923: Denote $F_{ab}$ as a total flux across the boundary between the cells:
924: \begin{equation}
925: F_{ab}\Delta t=M_a^{(R)}-M_b^{(R+h)}. \label{eq10a}
926: \end{equation}
927: If we take the Taylor expansion of Eq.~\eqref{eq9a} around $R$
928: and leave only the linear terms in $h$ we get:
929: \begin{equation}
930: F_{ab} \Delta t = \dfrac{h}{2}\left(\dfrac{1}{2}\dfrac{M}{R}-\dfrac{\partial M}{\partial R}\right).
931: \label{eq11a}
932: \end{equation}
933: It is convenient to rewrite the last equation in the form:
934: \begin{equation}
935: F_{ab} \Delta t = - \dfrac{1}{2}h R^{1/2} \dfrac{\partial}{\partial R}(MR^{-1/2}).
936: \label{eq12a}
937: \end{equation}
938:
939:
940: \section*{Appendix A3}
941: Let us comment on an important point in the above derivation. One may argue that
942: the factor 1/2 in Eq.~\eqref{eq14} is an arbitrary number. Thus, the factor
943: which appears in the right part of final Eq.~\eqref{eq-lbp-diff}
944: (now it is 1) is also arbitrary. However, we support our choice
945: of 1/2 in Eq.~\eqref{eq14} based on the following arguments. It is easy to rewrite
946: the derivation for 1D planar geometry i.e. to obtain the diffusion
947: equation in Cartesian coordinates. In terms of the above equations we should simply
948: set $M_a=M_b=M/2$ instead of using Eqs.~\eqref{eq6}--\eqref{eq7}.
949: The Eq.~\eqref{eq12} is replaced by
950: \begin{equation}
951: F_{ab} \Delta t = -\dfrac{h}{2}\dfrac{\partial M}{\partial R},
952: \end{equation}
953: and Eq.~\eqref{eq13} is rewritten as
954: \begin{equation}
955: M = \Sigma \cdot V_{r,\mathrm{turb}} \cdot \Delta t,
956: \end{equation}
957: where $M$ is the total mass of the current cell.
958: We should substitute these relations into analogue of Eq.~\eqref{eq17}
959: for planar geometry
960: \begin{equation}
961: \Delta R \cdot \Delta \Sigma = - (F_{ab}(R+h)-F_{ab}(R))\Delta t.
962: \end{equation}
963: In this way we obtain the classical equation:
964: \begin{equation}\label{eq19}
965: \frac{\partial\Sigma}{\partial t} = D \frac{\partial^2\Sigma}{\partial R^2},
966: \end{equation}
967: only if we introduce the diffusion coefficient, $D$, in the form of Eq.~\eqref{eq14}.
968: In other words, we choose the convention Eq.~\eqref{eq14} in such a way that
969: Eq.~\eqref{eq-lbp-diff} turns into Eq.~\eqref{eq19} in the limit of
970: $R \rightarrow \infty$ and $D$=const.
971:
972:
973: \bibliographystyle{apj.bst}
974: \bibliography{ms.bbl}
975: %\bibliography{/Users/dullemon/science/papers/citations/citations.bib}
976:
977: \end{document}
978: