1: %\documentclass[12pt,referee]{aastex}
2: \documentclass[12pt,preprint]{aastex}
3: %\usepackage{emulateapj5}
4: %\usepackage{epsf}
5:
6: %======================================================================
7: % author's own macros
8: %======================================================================
9:
10: \newcommand{\sect}[1]{\S\,\ref{#1}}
11: \newcommand{\be}{\begin{displaymath}}
12: \newcommand{\ee}{\end{displaymath}}
13: \newcommand{\bea}{\begin{eqnarray}}
14: \newcommand{\eea}{\end{eqnarray}}
15: \newcommand\msol{M_{\odot}}
16: \newcommand\cc{$^{12}$C/$^{13}$C}
17: %======================================================================
18:
19: %\received{}
20: %\accepted{}
21: %\journalid{}{}
22: %\articleid{}{}
23:
24: %\slugcomment{Astrophysical Journal, submitted}
25:
26: \shortauthors{Denissenkov \& Pinsonneault}
27: \shorttitle{$^3$He-Driven Mixing In Low-Mass Red Giants}
28:
29: %\received{2003 April 3}
30: \begin{document}
31:
32: \title{$^3$HE-DRIVEN MIXING IN LOW-MASS RED GIANTS:
33: CONVECTIVE INSTABILITY IN RADIATIVE AND ADIABATIC LIMITS}
34:
35: \author{Pavel A. Denissenkov\altaffilmark{1,2}, and Marc Pinsonneault\altaffilmark{1}}
36: \altaffiltext{1}{Department of Astronomy, The Ohio State University, 4055 McPherson Laboratory,
37: 140 West 18th Avenue, Columbus, OH 43210; dpa@astronomy.ohio-state.edu, pinsono@astronomy.ohio-state.edu.}
38: \altaffiltext{2}{On leave from Sobolev Astronomical Institute of St. Petersburg State University,
39: Universitetsky Pr. 28, Petrodvorets, 198504 St. Petersburg, Russia.}
40:
41: \begin{abstract}
42: We examine the stability and observational consequences of
43: mixing induced by $^3$He burning in the envelopes of first ascent red giants.
44: We demonstrate that there are two unstable
45: modes: a rapid, nearly adiabatic mode that we cannot identify with an underlying
46: physical mechanism, and
47: a slow, nearly radiative mode that can be identified with thermohaline convection.
48: We present observational constraints that make the operation
49: of the rapid mode unlikely to occur in real stars. Thermohaline convection turns out to be fast enough
50: only if fluid elements have finger-like structures with a length to diameter
51: ratio $l/d\ga 10$. We identify some potentially serious obstacles for thermohaline
52: convection as the predominant mixing mechanism for giants.
53: We show that rotation-induced horizontal turbulent diffusion
54: may suppress the $^3$He-driven thermohaline convection. Another potentially serious
55: problem for it is to explain observational evidence of enhanced extra mixing.
56: The $^3$He exhaustion in stars approaching the red giant branch (RGB) tip
57: should make the $^3$He mixing inefficient on the asymptotic giant branch (AGB).
58: In spite of this, there are observational data indicating the presence of extra mixing
59: in low-mass AGB stars similar to that operating on the RGB.
60: Overmixing may also occur in carbon-enhanced metal-poor stars.
61: \end{abstract}
62:
63: \keywords{stars: abundances --- stars: evolution --- stars: interiors}
64:
65: \section{Introduction}
66: \label{sec:intro}
67:
68: There is strong observational evidence for deep mixing in the radiative envelopes of
69: low-mass ($M\la 2\,\msol$) red giant branch (hereafter, LM-RGB) stars.
70: Changes in light element abundances (such as Li, C, N) and in the \cc\ ratio as
71: a function of luminosity have been seen in low- and solar-metallicity red giants both
72: in the field and in stellar clusters (e.g., \citealt{gb91,grea00,bea01,kea01,gea02,sh03,sm03,sb06,spiteea06}).
73: The observed pattern requires at least a component of {\it in situ} mixing.
74: This extra mixing could have consequences for other species (such as $^3$He) that are not directly observed.
75: Indeed, in spite of the predicted efficient production of $^3$He in low-mass main sequence (MS) stars,
76: its Galactic abundance has been nearly constant since the epoch of Big Bang nucleosynthesis (e.g., \citealt{t98,bea02,vfea03}).
77: To explain this, semi-empirical stellar evolution models have shown that the carbon depletion due
78: to extra mixing in LM-RGB stars should unavoidably be accompanied by
79: a strong $^3$He destruction that counterbalances its production on the MS (\citealt{rea84,h95,ch95,wea96}).
80:
81: However, the true physical process that is responsible for mixing has resisted identification.
82: Rotationally induced mixing has been an implied underlying mechanism since the pioneering work of
83: \cite{sm79}, but there are serious difficulties in reconciling
84: the observed mixing pattern with theoretical predictions (\citealt{chea05,pea06}).
85: Moreover, it can be shown that rotational mixing solely dependent on the local angular velocity gradient
86: $q\equiv (\partial\ln\Omega/\partial\ln r)$ (e.g., shear mixing) will be self-quenching.
87: Indeed, the empirically constrained mixing rate
88: $v_{\rm mix}\ga 10^{-3}$\,--\,$10^{-2}$\,cm\,s$^{-1}$ (see \sect{sec:concl}) is faster (as it should be)
89: than the mass inflow rate $|\dot{r}|\leq 10^{-4}$\,cm\,s$^{-1}$ in the radiative zones of LM-RGB stars. Hence, if $v_{\rm mix}$ were
90: proportional to $|q|^n\ (n>0)$ then the flattening of the rotation profile
91: by the accompanying angular momentum redistribution would quench the mixing very quickly.
92:
93: A very different class of solution has recently been proposed by
94: \cite{eea06} (hereafter, EDL06; see also \citealt{eea07}).
95: While investigating the core He flash in a low-mass model star near the RGB tip with the
96: code {\tt Djehuty}\footnote{This is a 3D explicit hydrodynamics code with the time step constrained
97: by the Courant condition which can couple to a 1D stellar evolution code.},
98: \cite{dea06} noticed some gas motions in the radiative zone above the H-burning shell,
99: in addition to convective motions driven by $^4$He burning in the core. In their follow-up papers,
100: they have made a conclusion that these additional gas motions
101: are due to the Rayleigh-Taylor instability (RTI) driven by $^3$He burning. They have noted
102: that, unlike other nuclear reactions in stars, the reaction $^3$He\,($^3$He,\,2p)$^4$He
103: decreases the mean molecular weight $\mu$. EDL06 have argued
104: that, even though the decrease of $\mu$ is minute ($|\Delta\mu|\sim 10^{-4}$), it is the resulting local inversion of
105: the $\mu$-gradient $\nabla_\mu \equiv (d\ln\mu/d\ln\,P)<0$ that has led to
106: the RTI in their 3D simulations. They claim that velocities of
107: gas motions induced by the RTI are ``comparable to the velocity of the normal
108: convection'' ($v_{\rm c}\sim 10^5$\,cm\,s$^{-1}$) and that this is consistent with the simple heuristic estimate of
109: $v^2\sim gH_P(\Delta\mu/\mu)$, where $g$ and $H_P$ are the local gravity and pressure scale height.
110:
111: A $\mu$-gradient inversion from $^3$He burning is a predicted consequence of standard stellar evolution.
112: After a low-mass star has left the MS its convective envelope first deepens (the first dredge-up) and then
113: its bottom starts to recede. At the depth of its maximum inward penetration the convective envelope imprints
114: a discontinuity in the chemical composition profile. Later on, the advancing in mass H-burning shell
115: will erase this discontinuity. During this event the evolution of the star slows down, which produces
116: bumps in differential luminosity functions of stellar clusters, and the star itself makes a tiny
117: zigzag on the HRD diagram (Fig.~\ref{fig:f1}, upper panel). However, before the major H-burning shell
118: erases the composition discontinuity, a shell in which $^3$He burns down and which advances
119: in front of the H-burning shell will cross the discontinuity first. The $^3$He($^3$He,2p)$^4$He reaction
120: reduces the mean molecular weight locally. However, this reduction is so minute, $\Delta\mu\approx\mu^2\Delta X_3/6\approx -10^{-4}$
121: (here, $\mu\approx 0.6$, and $X_3$ is the $^3$He mass fraction that can reach up to a value of $\sim$\,$2\times 10^{-3}$
122: in the envelope of a low-mass star ascending the RGB), that it is not seen on the $\mu$-profile
123: until the $\mu$ depression will find itself in the chemically homogeneous region of the radiative zone previously occupied
124: by the convective envelope. This happens close to the bump luminosity (Fig.~\ref{fig:f1}).
125:
126: In this paper we investigate the $^3$He instability in more detail.
127: First of all, we note that the RTI would not be expected
128: in compressible, and hence stratified, stellar material.
129: However, a $\mu$-gradient inversion may trigger instabilities related to convection.
130: In section 2, we discuss general criteria for convective instability taking into consideration that,
131: over longer timescales, fluid elements can exchange heat with their surroundings, thus modifying the background temperature stratification.
132: We identify two families of solutions: a rapid mode with a nearly adiabatic
133: thermal structure and a slow mode with a nearly radiative structure. The EDL06 results appear
134: to correspond to the rapid mode,
135: while the slow mode can be identified as thermohaline convection (e.g., \citealt{v04}). On the basis of
136: previously published diffusion coefficient estimates it is likely to
137: be triggered in the envelopes of red giants. However, there are significant (and
138: uncertain) assumptions related to the actual operation of this instability. We discuss
139: key features determining the efficiency of the slow mode: the assumed geometry of the fluid elements,
140: which directly impacts the timescale for exchanging heat; the potential impact of horizontal turbulence
141: in suppressing the instability; and the predicted depth of mixing.
142: In section 3 we demonstrate that the alternative approach based on the linear stability analysis of
143: the underlying conservation equations also leads to the conclusion that thermohaline convection
144: may be suppressed by the horizontal turbulent diffusion.
145:
146: In section 4 we evaluate the impact of any proposed mixing mechanism on the thermal
147: structure of the red giant branch stars. We argue that the rapid mode, which is similar in its properties to the originally
148: published results of EDL06, can be ruled out because it would induce strong feedback on the thermal
149: structure of giants and would predict a mixing pattern contrary to observations. We also demonstrate
150: that both the slow mode and prior empirical mixing estimates would not disturb the thermal structure of
151: giants. In section 4 we assess the overall promise of $^3$He mixing as a mechanism. We find that it may
152: be an attractive solution, but identify several potentially serious drawbacks on both observational
153: and theoretical grounds. In particular, we argue that previously published estimates of horizontal turbulence
154: would be sufficient to suppress the instability and that the naturally expected depth of mixing and trends
155: with luminosity may be in conflict with observational trends.
156:
157: \section{General Criteria for Convective Instability}
158:
159: In the presence of a positive $\mu$-gradient $d\mu/dr$ (a negative $\nabla_\mu$),
160: a fluid element displaced vertically upwards
161: will find itself surrounded by material with a higher $\mu$. Whether the fluid element will continue
162: to rise depends on how efficiently it can exchange mass and heat with its surroundings.
163: It is easy to anticipate that heat diffusion will favor the instability
164: by reducing the temperature gradient inside the rising fluid element $\nabla'\equiv (d\ln T'/d\ln P)$,
165: as compared to its adiabatic value $\nabla_{\rm ad}$, and slightly increasing the gradient in the surrounding
166: medium $\nabla\equiv (d\ln T/d\ln P)$, as compared to its radiative value $\nabla_{\rm rad}$ in the absence of mixing.
167: In a radiative zone, we always have $\nabla_{\rm rad}\leq \nabla < \nabla' < \nabla_{\rm ad}$. On the other hand,
168: molecular diffusion and horizontal turbulent diffusion (if the latter is present it will also contribute to
169: heat diffusion) will decrease the $\mu$-contrast between the fluid element and its
170: surroundings $|\Delta\mu| = |\mu' (r)-\mu (r)|$, thus hindering the development of convection.
171: Current theories of rotational mixing predict the existence of strong horizontal turbulence
172: (\citealt{chz92,m03,mpz04}); even if the vertical turbulence is
173: not strong enough to produce mixing such a horizontal turbulence could strongly impact
174: an instability driven by differences in composition.
175: Let us carry out a general investigation of effects produced by these diffusion processes
176: on the convective instability of the radiative zone in the presence of a $\mu$-gradient inversion.
177: The magnitude of these effects depends on the P\'{e}clet number that represents
178: a ratio of thermal and mixing time scale
179: $$
180: Pe = \frac{\tau_{\rm th}}{\tau_{\rm mix}}\approx\frac{lv}{(K+D_{\rm h})}\left(\frac{d}{l}\right)^2,
181: $$
182: where $\tau_{\rm th}\approx d^2/(K+D_{\rm h})$ and $\tau_{\rm mix} \approx l/v$.
183: Here, we consider a possibility that the convective motions
184: may be organized in elongated narrow structures resembling ``fingers'' whose length $l$ is much larger than their
185: diameter $d$. In this case, the fluid element velocity $v$, thermal diffusivity $K$, and
186: horizontal turbulent diffusivity $D_{\rm h}$ should appropriately be averaged over the finger length scale $l$.
187: For simplicity, we will represent a ``finger'' by a spherical fluid element of the diameter $d$
188: that travels the path $l\geq d$ before it dissolves. Modeling a ``finger'' with a cylinder would
189: introduce factors of order unity in our derived relations.
190: In the limit of a high P\'{e}clet number, mixing is so fast compared to heat exchange that $T'$ undergoes
191: nearly adiabatic changes, hence $\nabla'\approx\nabla_{\rm ad}$. In this limit, when the rising fluid element
192: dissolves it has a lower temperature than its surroundings therefore it reduces $T$ locally making $\nabla$ steeper
193: and closer to $\nabla_{\rm ad}$. In the opposite limit of a low P\'{e}clet number, mixing is so slow that
194: the radiative and turbulent heat flux from the surroundings can warm up the fluid element thoroughly as it rises.
195: This brings $\nabla'$ close to $\nabla_{\rm rad}$ while $\nabla$ remains nearly equal to $\nabla_{\rm rad}$
196: because the fluid element will have $T'\approx T$ when it dissolves.
197:
198: In our further analysis we will use simple heuristic relations obtained in the mixing length approximation
199: by \cite{m95} and \cite{tz97} in their investigations of shear instabilities in rotating stars.
200: We admit that the radiative zone may host both the $^3$He-driven convection and some other extra mixing of a nonconvective origin
201: (e.g., rotation-driven turbulent diffusion or meridional circulation) at the same time.
202: Following \cite{z92}, we assume that rotation-driven turbulence
203: is highly anisotropic, with horizontal components of the turbulent viscosity
204: strongly dominating over those in the vertical direction, $D_{\rm h}\gg D_{\rm v}$.
205: We consider a process with a diffusion coefficient $D_{\rm mix} = vl/3$. At the present stage we leave
206: $D_{\rm mix}$ unspecified, and simply solve for the coupled diffusion of heat
207: and chemicals to evaluate the range of diffusion coefficients over which an instability occurs.
208: In the next section we compare with specific (and previously
209: published) estimates of diffusion coefficients.
210:
211: The degree to which convection modifies the thermal
212: structure of the radiative zone depends on its heat transport efficiency
213: \bea
214: \Gamma = \frac{\Delta E_{\rm trans}}{\Delta E_{\rm ex}} = \frac{\nabla'-\nabla}{\nabla_{\rm ad}-\nabla'} =
215: \frac{D_{\rm mix}}{2(K+D_{\rm h})}\left(\frac{d}{l}\right)^2 = \frac{Pe}{6}.
216: \label{eq:gamma}
217: \eea
218: The quantity $\Gamma$ measures the deficit or excess of energy transported by rising or sinking turbulent
219: elements $\Delta E_{\rm trans}$ with respect to the energy $\Delta E_{\rm ex}$
220: the elements gain or lose through the radiative ($K$)
221: plus turbulent ($D_{\rm h}$) heat exchange with the surroundings.
222:
223: For an ideal gas, the square of the Brunt-V\"{a}is\"{a}l\"{a} (buoyancy) frequency is
224: \bea
225: N^2 = \frac{g}{H_P}\,(\nabla' - \nabla + \nabla_\mu - \nabla'_\mu).
226: \label{eq:n2}
227: \eea
228: This equation takes into account that horizontal diffusion (molecular plus turbulent) may change
229: the mean molecular weight of the fluid element $\mu'$ during its motion, so that $\nabla'_\mu\not= 0$.
230: The convective instability sets in when $N^2$ becomes negative.
231: By analogy with equation (\ref{eq:gamma}) and following \cite{tz97}, we introduce a $\mu$-transport efficiency
232: \bea
233: \Gamma_\mu = - \frac{\nabla'_\mu - \nabla_\mu}{\nabla'_\mu} = \frac{3D_{\rm mix}}{2(\nu_{\rm mol} +D_{\rm h})}\left(\frac{d}{l}\right)^2,
234: \label{eq:gammamu}
235: \eea
236: where $\nu_{\rm mol}$ is the molecular diffusivity.
237: Supplementing the radiation luminosity with a contribution to heat transport by convection
238: in the same manner as \cite{m95} dealt with shear mixing,
239: we obtain the following relation between $\nabla$, $\nabla_{\rm rad}$ and $\nabla_{\rm ad}$:
240: \bea
241: \nabla = \frac{\nabla_{\rm rad} + 6\frac{\Gamma^2}{\Gamma +1}\frac{K+D_{\rm h}}{K}\left(\frac{l}{d}\right)^2\nabla_{\rm ad}}
242: {1+6\frac{\Gamma^2}{\Gamma +1}\frac{K+D_{\rm h}}{K}\left(\frac{l}{d}\right)^2}.
243: \label{eq:nablas}
244: \eea
245:
246: Combining eqs. (\ref{eq:gamma}\,--\,\ref{eq:nablas}), the instability condition $N^2 < 0$ can be transformed into
247: \bea
248: -\nabla_\mu > \left(\frac{\Gamma_\mu + 1}{\Gamma_\mu}\right)
249: \frac{\Gamma}{1+\Gamma+6\Gamma^2\,\frac{K+D_{\rm h}}{K}\left(\frac{l}{d}\right)^2}\,\,(\nabla_{\rm ad} - \nabla_{\rm rad}).
250: \label{eq:conv}
251: \eea
252:
253: \subsection{The Adiabatic Limit}
254:
255: In the case of $\Gamma\gg 1$, equations (\ref{eq:gamma}) and (\ref{eq:nablas}) give
256: $\nabla'\approx\nabla\approx\nabla_{\rm ad}$, which we call ``the adiabatic limit''.
257: From equations (\ref{eq:gamma}) and (\ref{eq:gammamu}) it follows that
258: \bea
259: \Gamma_\mu = 3\,\frac{K+D_{\rm h}}{\nu_{\rm mol} +D_{\rm h}}\,\Gamma,
260: \label{eq:relgammas}
261: \eea
262: hence $\Gamma_\mu\gg 1$ as soon as $\Gamma\gg 1$ because $\nu_{\rm mol}\ll K$ (Fig.~\ref{fig:f3}).
263: Using these constraints, the condition (\ref{eq:conv}) is transformed to
264: \bea
265: D_{\rm mix} > \frac{1}{3}\,K\,\frac{\nabla_{\rm ad}-\nabla_{\rm rad}}{|\nabla_\mu|}\gg K.
266: \label{eq:adiab}
267: \eea
268: Radial displacements $l\sim 10^8$ cm of fluid elements with velocities of order $5\times 10^4$ cm\,s$^{-1}$
269: observed by EDL06 in their 3D red giant model above the major H-burning shell correspond to
270: a diffusion coefficient $D_{\rm mix}\sim\frac{1}{3}vl\sim 2\times 10^{12}$ cm$^2$\,s$^{-1} \gg K$.
271: The same or even higher order of magnitude estimate can be obtained if we calculate
272: $D_{\rm mix}\sim vH_P$, where $v^2\sim gH_P\Delta\mu/\mu$, as proposed by EDL06 (see next section).
273: So, it appears that the $^3$He-driven mixing in the 3D red giant model
274: of EDL06 somehow wound up in the metastable adiabatic limit.
275: We explore the consequences of such a rapid mixing process for both surface abundances and the thermal structure in section 4.
276:
277: \subsection{The Radiative Limit}
278:
279: Let us now consider a situation when $\Gamma(l/d)^2\ll 1$. This also means that $\Gamma\ll 1$ because we have assumed that $d\leq l$.
280: In this case, equations (\ref{eq:gamma}) and (\ref{eq:nablas}) give $\nabla'\approx\nabla\approx\nabla_{\rm rad}$, therefore
281: we will refer to it as ``the radiative limit''. Given that in equation (\ref{eq:relgammas})
282: the ratio $K/\nu_{\rm mol}\gg 1$ (solid curve in Fig.~\ref{fig:f3}), the assumption that $\Gamma\ll 1$ does not automatically
283: lead to $\Gamma_\mu < 1$ unless $D_{\rm h}\ga K$. Hence, we have to consider the cases of $\Gamma_\mu\gg 1$ and
284: $\Gamma_\mu < 1$ separately.
285:
286: a). In the radiative limit ($\Gamma\ll 1$), values of $\Gamma_\mu\gg 1$ can be met only if $D_{\rm h}\ll K$.
287: Using these constraints, the condition (\ref{eq:conv}) is simplified to
288: \bea
289: D_{\rm mix} < \frac{2K}{\nabla_{\rm ad}-\nabla_{\rm rad}}\,|\nabla_\mu|\left(\frac{l}{d}\right)^2,
290: \label{eq:dmix}
291: \eea
292: The right-hand side of (\ref{eq:dmix}) adequately reproduces
293: both the diffusion coefficient for thermohaline convection derived by \cite{kea80}
294: \bea
295: D_{\rm Kipp} = \frac{3K}{\nabla_{\rm ad}-\nabla_{\rm rad}}\,|\nabla_\mu|,
296: \label{eq:dkipp}
297: \eea
298: who argued that $l$ should be of order $d$, and the rate of mixing by elongated narrow ``fingers'' ($l>d$)
299: advocated by \cite{u72}
300: \bea
301: D_{\rm Ulrich} = \frac{8}{3}\pi^2\,\frac{K}{\nabla_{\rm ad}-\nabla_{\rm rad}}\,|\nabla_\mu|\left(\frac{l}{d}\right)^2.
302: \label{eq:dulrich}
303: \eea
304: Thus, in the radiative limit with $D_{\rm h}\ll K$ we can readily identify mixing with thermohaline convection.
305: Substituting expressions (\ref{eq:dkipp}\,--\,\ref{eq:dulrich}) in place of $D_{\rm mix}$, we find that, for thermohaline convection
306: driven by the $^3$He burning,
307: \bea
308: \Gamma \approx \frac{vl}{6K}\,\left(\frac{d}{l}\right)^2 \approx
309: \frac{D_{\rm mix}}{2K}\,\left(\frac{d}{l}\right)^2 \approx
310: \frac{|\nabla_\mu|}{\nabla_{\rm ad}-\nabla_{\rm rad}} \ll 1,
311: \label{eq:gamma2}
312: \eea
313: as we assumed.
314:
315: b). If $D_{\rm h}$ is not negligibly small compared with $K$ then we sure have $D_{\rm h}\gg\nu_{\rm mol}$, and relation
316: (\ref{eq:relgammas}) can be re-written as
317: \bea
318: \Gamma_\mu = 3\,\frac{K+D_{\rm h}}{D_{\rm h}}\,\Gamma.
319: \label{eq:relgammas2}
320: \eea
321: For $\Gamma_\mu < 1$, the condition (\ref{eq:conv}) takes a form
322: \bea
323: \frac{|\nabla_\mu|}{\nabla_{\rm ad}-\nabla_{\rm rad}} > \frac{1}{3}\frac{D_{\rm h}}{K+D_{\rm h}}
324: \label{eq:gmlt1}
325: \eea
326: (compare it with condition 5 from \citealt{v04}).
327: In the radiative zone of an LM-RGB star, the left-hand side of (\ref{eq:gmlt1}) is of order $10^{-3}$ at most.
328: A profile of the quantity $K$ in the radiative zone of our $0.83\,M_\odot$ bump luminosity model
329: is plotted with dashed curve in Fig.~\ref{fig:f3}. Given that $K\la 10^9$ cm$^2$\,s$^{-1}$ for
330: $r\la 0.1\,R_\odot$ and comparing the ratio $D_{\rm h}/K$ with the number $10^{-3}$,
331: we conclude that the horizontal turbulent diffusion
332: with $D_{\rm h} \ga 10^6$ cm$^2$\,s$^{-1}$ may hinder the development of convective instability.
333: Interestingly that such values of $D_{\rm h}$ have been found by \cite{pea06} in their low-metallicity
334: $0.85\,M_\odot$ bump luminosity model even for the less favorable case of solid-body rotation
335: of the convective envelope. For the case of differential rotation of the convective envelope,
336: which is suggested by observed fast rotation of horizontal branch stars, they have obtained
337: $D_{\rm h} \ga 10^7$ cm$^2$\,s$^{-1}$. For such large values of $D_{\rm h}$, the right-hand side of (\ref{eq:gmlt1})
338: exceeds the expression on the left-hand side at least for $r\la 0.1\,R_\odot$ (Fig.~\ref{fig:f3}),
339: therefore the convection (thermohaline mixing) will probably be suppressed there.
340:
341: \section{Ulrich's Solution in the Presence of Horizontal Turbulent Diffusion}
342: \label{sec:ulrich}
343:
344: In this section we demonstrate that \cite{u72} could have come to a conclusion about the suppression
345: of thermohaline convection by the strong horizontal turbulent diffusion similar to that made by us
346: (condition \ref{eq:gmlt1}) if he had included $D_{\rm h}$ in his equations. To do this, we start with
347: the linearized equations of conservation of momentum, thermal energy, and chemical composition
348: similar to those used by Ulrich (his equations 1\,--\,3) but with the diffusion coefficient
349: $D_{\rm h}$ taken into account
350: \bea
351: \label{eq:ulrich1}
352: 2\,\frac{dv}{dt} & = & -\frac{\nu}{d^2}\,v - g\,\delta\ln T + g\,\delta\ln\mu, \\
353: \label{eq:ulrich2}
354: \frac{d\,\delta\ln T}{dt} & = & \frac{(\nabla_{\rm ad}-\nabla_{\rm rad})}{H_P}\,v - \frac{(K+D_{\rm h})}{d^2}\,\delta\ln T, \\
355: \label{eq:ulrich3}
356: \frac{d\,\delta\ln\mu}{dt} & = & -\frac{\nabla_\mu}{H_P}\,v - \frac{(\nu_{\rm mol}+D_{\rm h})}{d^2}\,\delta\ln\mu. \\ \nonumber
357: \eea
358: Here, $\nu = \nu_{\rm mol} + \nu_{\rm rad}$ is the total (molecular plus radiative) viscosity,
359: $\delta\ln T = \ln T'(r) - \ln T(r)$, $\delta\ln\mu = \ln\mu'(r) - \ln\mu (r)$, other symbols having been
360: defined previously.
361:
362: The characteristic polynomial for the linear system of ODEs (\ref{eq:ulrich1}\,--\,\ref{eq:ulrich3})
363: can be written in the following form:
364: \bea
365: \label{eq:dispersion}
366: \left(\omega\tau_{\rm th}\right)^3 + \left[1+\left(\tau_{\rm th}/\tau_\nu\right)+\left(\tau_{\rm th}/\tau_\mu\right)\right]\,\left(\omega\tau_{\rm th}\right)^2 + & \\ \nonumber
367: \left[\left(\tau_{\rm th}/\tau_\nu\right)+\left(\tau_{\rm th}/\tau_\nu\right)\left(\tau_{\rm th}/\tau_\mu\right)+
368: \left(\tau_{\rm th}/\tau_\mu\right)+\left(\tau_{\rm th}^2N^2\right)\right]\,\left(\omega\tau_{\rm th}\right) + & \\ \nonumber
369: \left[\left(\tau_{\rm th}/\tau_\nu\right)\left(\tau_{\rm th}/\tau_\mu\right)+\left(\tau_{\rm th}^2N_\mu^2\right)+
370: \left(\tau_{\rm th}/\tau_\mu\right)\left(\tau_{\rm th}^2N_T^2\right)\right] = 0, & \\ \nonumber
371: \eea
372: where $\omega = 2\pi\tau_{\rm mix}^{-1}$ is the eigen frequency of stable ($\omega < 0$) or unstable ($\omega > 0$) mode,
373: $N_T^2 = g(\nabla_{\rm ad}-\nabla_{\rm rad})\,H_P^{-1}$ and $N_\mu^2 = g\nabla_\mu\,H_P^{-1}$ are the squares of the $T$- and $\mu$-component of
374: the Brunt-V\"{a}is\"{a}l\"{a} frequency for the ideal gas, $N^2 = N_T^2 + N_\mu^2$, while
375: $\tau_\nu = d^2/\nu$, $\tau_{\rm th} = d^2/(K+D_{\rm h})$, and $\tau_\mu = d^2/(\nu_{\rm mol}+D_{\rm h})$
376: denote the viscous, thermal, and horizontal diffusion timescales for the fluid element.
377: Our equation (\ref{eq:dispersion}) is equivalent to Ulrich's dispersion relation (10).
378:
379: The only real root of the polynomial (\ref{eq:dispersion}) is plotted as a function of $(D_{\rm h}/K)$ in Fig.~\ref{fig:f6} for
380: the ratio $d/H_P = 0.01$ and three different values of $N_\mu^2 = -10^{-7}$ (solid line), $-10^{-6}$ (short-dashed line),
381: and $-10^{-8}$ (long-dashed line). The first value of $N_\mu^2$ is close to the minimum one found in the region of the
382: $\mu$-gradient inversion produced by $^3$He burning in a low-metallicity bump luminosity star with a mass $M\approx 0.8\,\msol$. The quantities $\left(\omega\tau_{\rm th}\right)$
383: and $(D_{\rm h}/K)$ in Fig.~\ref{fig:f6} have been scaled appropriately to reveal both our guessed functional dependence (\ref{eq:dulrich}) and
384: instability condition (\ref{eq:gmlt1}). At a fixed value of $N_\mu^2$ our solutions for the ratio $d/H_P$ varying from
385: $0.01$ down to $0.0001$ coincide. From Fig.~\ref{fig:f6}, we conjecture that
386: \bea
387: D_{\rm mix} \sim \frac{l^2}{\tau_{\rm mix}} \sim \frac{l^2}{\tau_{\rm th}}\,\left(\omega\tau_{\rm th}\right) \sim K\,\frac{|N_\mu^2|}{N_T^2}\,\left(\frac{l}{d}\right)^2\times
388: \left(1 - \frac{D_{\rm h}}{K}\,\frac{N_T^2}{|N_\mu^2|}\right),
389: \label{eq:modulrich}
390: \eea
391: i.e. the thermohaline convective instability may develop ($\omega > 0$) only if $D_{\rm h} < K|N_\mu^2|/N_T^2$. The latter condition is equivalent
392: (ignoring factors of order unity) to (\ref{eq:gmlt1}) for $D_{\rm h} < K$. If $D_{\rm h}\ll K$ then we can neglect the term in the parentheses.
393: In this case we obtain an expression for $D_{\rm mix}$ similar to Ulrich's original formula.
394:
395: Note that an equation similar to (\ref{eq:modulrich}) can be derived directly from the dispersion relation
396: (\ref{eq:dispersion}) in the same way that Ulrich used to estimate $D_{\rm mix}$ in (\ref{eq:dulrich}).
397: Following him, we ignore the viscosity and take advantage of the fact that the thermohaline mode is
398: the slowest one ($\omega\tau_{\rm th}\sim\tau_{\rm th}/\tau_{\rm mix}\ll 1$).
399: Therefore, we can neglect the quadratic and cubic terms in (\ref{eq:dispersion}) as well as terms with $\tau_\nu^{-1}$.
400: Taking into account that $N_T^2\gg |N_\mu^2|$, and $N^2 \gg \left(\tau_{\rm th}\tau_\mu\right)^{-1}$, we finally obtain
401: \bea
402: D_{\rm mix} \sim (K+D_{\rm h})\,\frac{|N_\mu^2|}{N_T^2}\,\left(\frac{l}{d}\right)^2\times
403: \left[1 - \frac{(\nu_{\rm mol}+D_{\rm h})}{(K+D_{\rm h})}\,\frac{N_T^2}{|N_\mu^2|}\right],
404: \label{eq:modulrich2}
405: \eea
406: which is reduced to (\ref{eq:modulrich}) if $\nu_{\rm mol} < D_{\rm h} < K$.
407:
408: \section{Observational Constraints on the Extra-Mixing Rate}
409: \label{sec:mixing}
410:
411: The $^3$He mechanism has a different underlying origin than rotational mixing,
412: and it is therefore useful to re-evaluate the global implications of its operation.
413: This is especially true because we have identified two stable branches with very
414: different timescales. We begin by establishing that a mixing process which occurs over
415: too short of a timescale would have a dramatic impact on the thermal structure which contradicts
416: the observational data. We also demonstrate that diffusion coefficients derived from empirical
417: mixing estimates are consistent with the data.
418:
419: We then critically examine whether thermohaline
420: mixing is capable of reproducing the data, and the answer depends critically on the assumed geometry
421: of the fluid elements (and the rate at which they can achieve thermal balance with their surroundings).
422: However, we can also extend the same underlying mechanism to predict luminosity trends, behavior in
423: other evolutionary states, and implications for interacting binaries. Definite predictions emerge, and
424: we can identify both existing conflicts and topics which will require further calculation.
425: In particular, we contend that the natural expectation would be for mixing that is shallower
426: in temperature and weaker for bright giants than current data indicates. The predictions for other types
427: of stars are fundamentally different, and in our view the latter test will ultimately prove to be decisive.
428:
429: \subsection{Rapid Mixing in the Adiabatic Limit}
430:
431: The observed changes in the surface abundances of Li and C and in the \cc\ ratio
432: as a function of luminosity in LM-RGB stars (e.g., \citealt{chea98,grea00}) can be used to constrain the depth and rate of
433: extra mixing in them. If $\Delta\log T$ is the difference between the logarithms of temperature at the base of the
434: H-burning shell and at the maximum depth of extra mixing and $D_{\rm mix}$ is the diffusion coefficient then,
435: as \cite{dv03} have demonstrated, extra mixing in LM-RGB stars can be
436: parameterized by any pair of correlated values within the close limits
437: specified by $\Delta\log T\approx 0.19$ and $D_{\rm mix}\approx 4\times 10^8$\,cm$^2$\,s$^{-1}$, to
438: $\Delta\log T\approx 0.22$ and $D_{\rm mix}\approx 8\times 10^8$\,cm$^2$\,s$^{-1}$.
439: However, that parameterization
440: did not take into consideration the thermal response of the radiative zone
441: to mixing.
442:
443: In this paper, we use a more consistent
444: parametric prescription by letting $D_{\rm mix}$ be equal to a fixed fraction of $K$
445: and allowing the temperature gradient in the radiative zone to be modified as prescribed by
446: the mixing length theory relation (\ref{eq:nablas}).
447: It is inspired by a similarity between
448: equation (\ref{eq:dkipp}) and the functional dependence of $D_{\rm mix}$ on $K$ obtained for
449: rotational shear mixing by \cite{mm96}.
450:
451: We derive $\Gamma = 0.01$ for the mixing depth $\Delta\log T=0.19$ constrained by \cite{dv03}, which results in
452: the diffusion coefficient $D_{\rm mix}=0.02\,K$ (assuming that $l\approx d$ and $D_{\rm h}\ll K$ in equation \ref{eq:gamma}).
453: These models reproduce quite well the chemistry of LM-RGB stars
454: above the bump luminosity (dashed curves in Fig.~\ref{fig:f2}). They do not
455: noticeably change our models' photometric behavior and evolutionary time scale near the bump luminosity
456: compared to models without extra mixing. This is important because
457: photometric observations of LM-RGB stars, in particular the absolute V-band magnitude of the luminosity bump
458: and the excess number of stars in the bump, agree with predictions of standard stellar evolution
459: theory (e.g., \citealt{bch06}). Our test computations have shown that the ratio of the gradients
460: in equations (\ref{eq:dkipp}\,--\,\ref{eq:dulrich}) does not change much with radius
461: and that it is roughly proportional to the abundance of $^3$He left in the mixing zone.
462:
463: On the other hand, already at $\Gamma = 0.4$ the bump luminosity
464: zigzag gets so extended toward a lower luminosity (dashed curve in upper panel in Fig.~\ref{fig:f1}), and the model star spends
465: so long time following it that this peculiar behavior would sure have been noticed in
466: photometric studies, such as the counting of the number densities of stars as a function of luminosity on the RGBs of
467: globular clusters. To be more specific, it takes about twice as long for the model star
468: to make the extended zigzag as compared to the standard evolution. It should also be noted
469: that the model of such a rapidly mixed star spends most of this time residing near the bottom of the zigzag, about
470: 0.3 magnitude below the standard bump luminosity. Values of $\Gamma > 0.4$ would result in
471: even more drastic changes. This behavior is similar to that
472: of models of rapidly rotating RS CVn binaries found by \cite{denea06} except that
473: in the latter case the extended zigzag was produced by an increase of $\nabla$ caused
474: by the stars' rotational deformations.
475:
476: The estimate of the turbulent velocity $v^2\sim gH_P(\Delta\mu/\mu)$ for the $^3$He-driven mixing
477: used by EDL06 can be obtained from equation (\ref{eq:n2}) if we put into it $\nabla' - \nabla = \nabla'_\mu = 0$.
478: Indeed, in this case we can approximately consider that
479: $v^2\sim H_P^2\,|N^2|=gH_P|\nabla_\mu|=gH_P^2(d\ln\mu/dr)\sim gH_P(\Delta\mu/\mu)$, at least in the vicinity of
480: the $^3$He-burning shell where the mean molecular weight height scale is of order $H_P$ (bottom panel in Fig.~\ref{fig:f1}).
481: Note that in the mixing length theory the approximations $\nabla' = \nabla$
482: and $\nabla'_\mu = 0$ are correct only in the limits of $\Gamma = \infty$
483: or $\Gamma = 0$, and $\Gamma_\mu = \infty$ (eqs. \ref{eq:gamma}\,--\,\ref{eq:nablas}).
484:
485: Neglecting the influence of extra mixing on the radiative zone's thermal stratification,
486: we have computed the evolution of our model also with the following diffusion coefficient:
487: \bea
488: D_{\rm mix}=\frac{1}{3}H_Pv,\ \ \mbox{where}\ \ v^2=\frac{1}{8}gH_P|\nabla_\mu|.
489: \label{eq:dmu}
490: \eea
491: The factor $\frac{1}{8}$ comes from the mixing length theory (\citealt{wea04}, Ch.~14).
492: The depth of this ``$^3$He-driven'' extra mixing has been determined by locating the minimum of $\mu$
493: above the H-burning shell. Outside of this point, $\mu$ increases with radius due to the $^3$He burning and
494: mixing. We think that our prescription is in line with that EDL06 had in mind.
495: It should be noted that in our computations values of $D_{\rm mix}$ were determined at each time step
496: using a current distribution of $\mu$ that was constantly modified by extra mixing.
497: Characteristic values of $D_{\rm mix}$ obtained in this self-regulating way were of order $10^{12}$\,cm$^2$\,s$^{-1}$.
498: Such fast extra mixing is known to produce large amounts of $^7$Li via the Cameron-Fowler
499: mechanism (e.g., \citealt{dw00,dh04}). This disagrees with the low (often undetectable) Li abundances in the majority of
500: LM-RGB stars located above the bump luminosity (compare the solid curve above $\log\,L/L_\odot\approx 1.8$
501: with the observational data points in top panel in Fig.~\ref{fig:f2}).
502: In accordance with EDL06, we did find a modest decrease in the \cc\ ratio.
503: However, it is obvious that the observed evolutionary changes of the surface chemical composition
504: of LM-RGB stars require a slightly deeper (in order to reproduce the C depletion)
505: and much slower (in order to keep the Li abundance low) extra mixing than that advocated by EDL06.
506: Besides, extra mixing with diffusion coefficients $D_{\rm mix}\gg K\sim 10^8$\,--\,$10^{10}$\,cm$^2$\,s$^{-1}$
507: would bring the radiative zone to the quasi-adiabatic state (unless the fluid elements
508: have finger-like structures, which has not been reported by EDL06), which would cause the star to make
509: a prolonged excursion below the bump luminosity in contradiction with observations.
510: Therefore, we believe that the rapid mode originally invoked by EDL06 does not operate,
511: but that mild mixing is indicated.
512:
513: \subsection{Mild Thermohaline Mixing in the Radiative Limit}
514:
515: Because the $^3$He-driven thermohaline convection is expected to work in the radiative limit it is
516: interesting to test if it is fast and deep enough to
517: explain extra mixing in LM-RGB stars. In Fig.~\ref{fig:f4}, we illustrate characteristic diffusion
518: coefficients.
519: In order to produce mixing, a physical mechanism must operate over a timescale shorter than
520: the inflow rate.
521: The dot-dashed curve shows our empirically constrained diffusion coefficient $D_{\rm mix}=0.02\,K$.
522: For comparison, the dashed curve shows a minimum threshold diffusion coefficient
523: $D_{\rm inflow}=|\dot{r}|H_P$, where $|\dot{r}|$ is a mass inflow rate of H-rich material
524: that flows from the bottom of convective envelope toward the H-burning shell (resembling a spherical
525: accretion).
526: The bottom solid curve shows a profile of
527: the diffusion coefficient (\ref{eq:dkipp}) in our unmixed bump luminosity model.
528: Once mixing ensues, the $\mu$-gradient inversion spreads out over the entire radiative zone
529: above the $^3$He-burning shell.
530: The final state is illustrated with the top solid line.
531: We conclude that, in the prescription given by \cite{kea80},
532: the thermohaline convection could marginally commence and mix a narrow region in the vicinity of the $^3$He-burning
533: shell (at $r\approx 0.08\,R_\odot$ in Fig.~\ref{fig:f4}). However, the Kippenhahn et al.
534: diffusion coefficient is only marginally large enough to trigger this process and is two orders of
535: magnitude below the empirical value.
536:
537: If we adopt the Ulrich prescription (\ref{eq:dulrich}) with
538: $l\approx 10\,d$, our diffusion coefficient is raised by a factor of $10^2$ and the mechanism
539: may be viable. This is the approach advocated by \cite{chz07} (incidentally, their paper
540: was posted on astro-ph on the same day when we submitted the first version of our paper).
541:
542: Unfortunately, \cite{chz07} do not explain how they have chosen the depth of mixing.
543: We find that the depth corresponding to a minimum on the $\mu$-profile (solid vertical line segments in Fig.~\ref{fig:f5})
544: is not sufficient to produce the observed C depletion (top second panel in Fig.~\ref{fig:f2}).
545: An overshooting on a length scale of order $H_P$ could do it (dotted vertical line
546: segments in Fig.~\ref{fig:f5} are placed at a distance $H_P$ below $\mu_{\rm min}$) but then
547: thermohaline ``fingers'' would have to penetrate down a region of higher $\mu$ where they
548: should experience a strong breaking. It should also be noted that the penetration of
549: a region with the negative $d\mu/dr$ below $\mu_{\rm min}$ would reduce the average mixing rate by
550: decreasing the slope of the positive $d\mu/dr$ in the mixing zone. Given these
551: uncertainties that cannot be resolved from first principles but instead require empirical
552: calibrations and/or higher resolution 3D hydrodynamic simulations,
553: we postpone the use of equation (\ref{eq:dulrich}) to our future paper.
554:
555: \section{Concluding Remarks}
556: \label{sec:concl}
557:
558: In this paper, we have shown that the $^3$He burning in the radiative zone of an LM-RGB star
559: may drive convective fluid motions
560: provided that their heat transport efficiency is either very high (the adiabatic limit) or
561: extremely low (the radiative limit).
562: Confirming the conclusions made by \cite{chz07},
563: we identify the mixing in the radiative limit with thermohaline convection and we
564: note that this convection would have a sufficiently high rate to explain the observed mixing pattern in LM-RGB stars
565: only if fluid elements could travel over length scales exceeding their diameters by a factor of 10 or more.
566: However, we also find that thermohaline convection may be suppressed by
567: horizontal turbulence if its associated diffusivity $D_{\rm h}\ga 3K|\nabla_\mu|/(\nabla_{\rm ad}-\nabla_{\rm rad})$.
568: Such values of $D_{\rm h}$ for rotation-induced horizontal turbulence have been obtained
569: by \cite{pea06} who used a prescription for estimating $D_{\rm h}$ proposed by \cite{mpz04}.
570: Although this prescription may be considered
571: as a very primitive approximation to a complex physical phenomenon,
572: a similar heuristic approach has been used
573: to successfully model mixing and angular momentum transport in radiative zones of massive MS stars (e.g., \citealt{tz97,tea97,m03}).
574: Of course, we recognize that rigorous 3D hydrodynamic simulations have yet to be done
575: to support or refute these heuristic models.
576: In addition to the theoretical issues above, there are significant
577: empirical challenges for an explanation that relies solely on thermohaline convection.
578: The same physics should consistently be applied to other phases of evolution or
579: situations where $\mu$ inversions occur.
580:
581: If extra mixing in RGB stars is really driven by $^3$He burning then it should
582: die out by the end of the RGB evolution because of the $^3$He exhaustion.
583: In this case, the $^3$He-driven extra mixing could not resume working
584: in the same stars on the asymptotic giant branch (AGB). So, we would expect
585: the absence of observational signatures of extra mixing in low-mass ($M\la 2\,M_\odot$)
586: AGB stars unless the mixing in them is of a different nature.
587: However, given the similarities in their depth and in the structure of radiative zone
588: where they operate, it is unlikely that the RGB and AGB mixing
589: have different physical mechanisms.
590: Contrary to this prediction, there are observational data
591: indicating the presence or necessity of operation of extra mixing in these stars
592: (e.g., \citealt{nbw03,mea06}).
593: Moreover, to comply with observations,
594: the AGB mixing has to penetrate close enough to the H-burning shell
595: to dredge up material processed in the CN-cycle, like in RGB stars, mimicking the convective
596: hot-bottom burning that occurs in more massive AGB stars.
597:
598: \cite{sea07} have used the Kippenhahn et al. prescription (\ref{eq:dkipp}) to model
599: thermohaline mixing in a metal-poor low-mass
600: MS star accreting wind material from its AGB binary companion enriched in He and C.
601: Such accretion is believed to be the primary process responsible for the formation of
602: the so-called carbon-enhanced metal-poor (CEMP) stars. They have found that thermohaline
603: convection mixes almost 90\% of the star within about $10^9$ years after the accretion.
604: On the other hand, the RGB mixing pattern can be reproduced only if the diffusion coefficient given by
605: equation (\ref{eq:dkipp}) is increased by a factor of $10^2$ to $10^3$ (\citealt{chz07}; also see our Fig.~\ref{fig:f4}).
606: In this case, thermohaline mixing in CEMP MS stars would dilute the accreted material
607: on a much shorter time scale of order $10^6$\,--\,$10^7$ years. Unless
608: most of the CEMP stars accreted substantial fractions of their initial masses,
609: their rather high frequency $f_{\rm CEMP}\ga$\,20\% among very metal-poor stars (e.g., \citealt{lea06})
610: would look surprising. Furthermore, both \cite{lea06} and \cite{aea07} have found
611: anti-correlations between [C/H] (and [(C+N)/H]) and luminosity for Ba-enhanced CEMP stras
612: spanning over three orders of magnitude in $L$ that they interpreted as
613: an evidence of dilution of the envelope material in the accreting companion.
614: Thermohaline convection on a time scale of order $10^6$\,--\,$10^7$ years
615: would mix the CEMP MS stars almost instantaneously and well before
616: their luminosity would begin to increase due to the core H exhaustion.
617: In that case, the mentioned anti-correlations could not have appeared.
618: Also note that even on the lower RGB Ba-enhanced CEMP stars have quite low carbon isotopic ratios
619: ($^{12}$C/$^{13}$C\,$ < 20$; \citealt{rea05}) in a striking contrast with the values of
620: $^{12}$C/$^{13}$C\,$ \gg 1000$ that the low-mass AGB stars are predicted to
621: return to the interstellar medium (\citealt{h04}). Extra mixing (in the low-mas AGB stars)
622: could easily resolve this discrepancy.
623:
624: The problem of mixing in CEMP MS and RGB stars has recently been addressed by \cite{dp07}.
625: Particularly, they have shown that the first dredge-up dilution of CN enrichment in CEMP stars relative to
626: their MS precursors is indeed a plausible explanation of the observed anticorrelation of [N/Fe]
627: with $\log\,L/L_\odot$ and that it contradicts models that rely on efficient thermohaline mixing induced by small $\mu$
628: gradients in red giants. This result has independently been confirmed by \cite{aea08}.
629: The suppression of thermohaline convection by rotationally driven horizontal turbulence
630: may explain its reduced efficiency in MS CEMP stars.
631:
632: Another potentially serious problem for the $^3$He-driven thermohaline convection
633: could be to explain available observational evidence of enhanced extra mixing in LM-RGB stars.
634: Firstly, observations show that
635: in some globular clusters the anti-correlated abundance variations of C and N in red giants
636: become larger when stars approach the RGB tip. Moreover, extremely large values of
637: the N abundance in some of these stars indicate the dredge-up of material in which not only C
638: but also a fraction of O has been converted into N (\citealt{sbh05}).
639: Secondly, at least in the globular cluster M13, the relative number of upper RGB stars with
640: the O--Na anti-correlation increases with luminosity (\citealt{jea05}).
641: Thirdly, \cite{dpt06} have shown that the $^{19}$F abundance variations found in bright
642: red giants of the globular cluster M4 by \cite{sea05} may also require that extra mixing in them to operate
643: much faster and somewhat deeper than in LM-RGB stars immediately above the bump luminosity.
644: It is difficult to interpret these data by the $^3$He-driven mixing because its efficiency should
645: decline toward the RGB tip in proportion as $^3$He gets depleted.
646: Unfortunately, the question of evolutionary Na, O, and $^{19}$F abundance variations in
647: globular-clusters RGB stars is still a matter of debate from the observational point of view.
648: Therefore, we consider them as a potential rather than a real problem
649: for the $^3$He thermohaline convection.
650:
651: A similar problem
652: is encountered when one tries to understand the phenomenon of Li-rich giants.
653: There are convincing arguments that high Li abundances in these LM-RGB stars are produced via the Cameron-Fowler
654: mechanism that also requires enhanced extra mixing with $D_{\rm mix}\sim 10^{10}$\,--\,$10^{11}$ cm$^2$\,s$^{-1}$
655: (\citealt{dw00,dh04}). It should be noted that most of the Li-rich giants are located above the bump luminosity
656: (\citealt{chb00}). Besides, their proportion among rapid rotators ($v\sin\,i\geq 8$\,km\,s$^{-1}$)
657: is $\sim$\,50\% which is considerably larger than $\sim$\,2\% of Li-rich stars among the much more common slowly
658: rotating ($v\sin\,i\la 1$\,km\,s$^{-1}$) K-giants (\citealt{dea02}). It is not clear why
659: the $^3$He-driven mixing would be enhanced in fast rotators. Oppositely, \cite{c99} argues
660: that a larger shear due to differential rotation should decrease the efficiency of mixing by thermohaline convection.
661: A larger shear would also intensify horizontal turbulence, thus hindering thermohaline convection even stronger.
662: So, the Li-rich giants seem to support the hypothesis of rotation-induced mixing rather than that of
663: thermohaline mixing. This may not necessarily be rotational shear mixing that has already been criticized.
664: Instead, rotation may deposit its kinetic energy to mixing less directly, e.g. through generation of
665: buoyant magnetic flux tubes (\citealt{bea07}).
666:
667: For the adiabatic limit, the predicted evolutionary changes of
668: the surface composition of LM-RGB stars disagree with observational data.
669: Besides, high $\Gamma$'s would bring the radiative zone to the quasi-adiabatic
670: state which would result in a photometric behavior of the RGB star inconsistent with the observed one.
671:
672: In the vicinity of $\mu_{\rm min}$, our empirically constrained diffusion coefficient
673: has values of order $D_{\rm mix}\sim 10^6$\,--\,$10^7$ cm$^2$\,s$^{-1}$ (dot-dashed curve in Fig.~\ref{fig:f4}).
674: If we assume that in real LM-RGB stars extra mixing is produced by thermohaline convection whose
675: fluid elements have a ratio of $l/d\sim 10$, where $l$ does not exceed
676: the local pressure scale height $H_P\sim 0.02\,R_\odot$, then we can estimate the elements' characteristic velocities
677: $v\approx 3D_{\rm mix}/l \sim 2\times 10^{-3}$\,--\,$2\times 10^{-2}\,(H_P/l)$ cm\,s$^{-1}$. EDL06 have found velocities of order
678: $5\times 10^4$ cm\,s$^{-1}$ in their 3D red giant model. Those would correspond to fluid elements with
679: diameters from 6 to 60 cm! Interestingly, such small fluid elements would actually be optically
680: thin because the photon mean free path is $\sim$\,1 cm in this environment. However, we do not believe
681: that EDL06 could resolve such small finger-like structures. As we mentioned, they have reported fluid element displacements
682: of order $l\sim 10^8$ cm. Given the discussed inconsistencies in modeling extra mixing in LM-RGB stars
683: with the $^3$He-driven convection, we conclude that a different mechanism is worth
684: searching for. It is also obvious that higher resolution 3D hydrodynamic simulations of the $^3$He-driven
685: mixing are needed to understand what EDL06 have actually witnessed. In particular,
686: we qoute here an issue raised by the referee, who seems to be an expert in the field.
687: ``The EDL06 calculations had no mechanism to simulate the turbulent cascade on scales smaller than the zoning
688: (a subgrid scale model, such as an eddy viscosity treatment), and yet the calculations did not numerically
689: blow up. To me this says that the finite difference expressions of the {\tt Djehuty} code are themselves
690: quite diffusive. Thus, I am surprised that this incredibly small mean molecular weight inversion could generate
691: significant motion without being squelched by the numerical diffusion. My conclusion is that there are some
692: serious issues that must be addressed about the numerical behavior of the 3D calculations.''
693:
694: \acknowledgements
695: We thank the anonymous referee for useful comments and suggestions
696: that helped us to improve the manuscript.
697: We acknowledge support from the NASA grant NNG05 GG20G.
698:
699: %\newpage
700:
701: \begin{thebibliography}{}
702:
703: \bibitem[Aoki et al.(2008)]{aea08}
704: Aoki, W., Beers, T.~C., Sivarani, T., Marsteller, B., Lee, Y.~S., Honda, S.,
705: Norris, J.~E., Ryan, S.~G., \& Carollo, D.~2008, arXiv:0801.4187v1 [astro-ph]
706:
707: \bibitem[Aoki et al.(2007)]{aea07}
708: Aoki, W., Beers, T.~C., Christlieb, N., Norris, J.~E., Ryan, S.~G., \& Tsangarides, S.~2007,
709: ApJ, 655, 492
710:
711: \bibitem[Bania, Rood, \& Balser(2002)]{bea02}
712: Bania, T.~M., Rood, R.~T., \& Balser, D.~S.~2002, Nature, 415, 54
713:
714: \bibitem[Bellman et al.(2001)]{bea01}
715: Bellman, S., Briley, M.~M., Smith, G.~H., \& Claver, C.~F.~2001,
716: PASP, 113, 326
717:
718: \bibitem[Bjork \& Chaboyer(2006)]{bch06}
719: Bjork, S.~R., \& Chaboyer, B.~2006, ApJ, 641, 1102
720:
721: \bibitem[Busso et al.(2007)]{bea07}
722: Busso, M., Wasserburg, G.~J., Nollett, K.~M., \& Calandra, A.~2007, ApJ, 671, 802
723:
724: \bibitem[Canuto(1999)]{c99}
725: Canuto, V.~M.~1999, ApJ, 524, 311
726:
727: \bibitem[Chaboyer \& Zahn(1992)]{chz92}
728: Chaboyer, B., \& Zahn, J.-P.~1992, A\&A, 253, 173
729:
730: \bibitem[Charbonnel(1995)]{ch95}
731: Charbonnel, C.~1995, ApJ, 453, L41
732:
733: \bibitem[Charbonnel, Brown, \& Wallerstein(1998)]{chea98}
734: Charbonnel, C., Brown, J.~A., \& Wallerstein, G.,~1998, A\&A, 332, 204
735:
736: \bibitem[Charbonnel \& Balachandran(2000)]{chb00}
737: Charbonnel, C., \& Balachandran, S.~C.~2000, A\&A, 359, 563
738:
739: \bibitem[Charbonnel \& Zahn(2007)]{chz07}
740: Charbonnel, C., \& Zahn, J.-P.~2007, A\&A, 467, L15
741:
742: \bibitem[Chanam\'{e}, Pinsonneault, \& Terndrup(2005)]{chea05}
743: Chanam\'{e}, J., Pinsonneault, M., \& Terndrup, D.~M.~2005, ApJ, 631, 540
744:
745: \bibitem[Dearborn, Lattanzio, \& Eggleton(2006)]{dea06}
746: Dearborn, D.~S.~P., Lattanzio, J.~C., \& Eggleton, P.~P.~2006, ApJ, 639, 405
747:
748: \bibitem[Denissenkov \& Weiss(2000)]{dw00}
749: Denissenkov, P.~A., \& Weiss, A.~2000, A\&A, 358, L49
750:
751: \bibitem[Denissenkov \& VandenBerg(2003)]{dv03}
752: Denissenkov, P.~A., \& VandenBerg, D.~A.~2003, ApJ, 593, 509
753:
754: \bibitem[Denissenkov \& Herwig(2004)]{dh04}
755: Denissenkov, P.~A., \& Herwig, F.~2004, ApJ, 612, 1081
756:
757: \bibitem[Denissenkov, Chaboyer, \& Li(2006)]{denea06}
758: Denissenkov, P.~A., Chaboyer, B., \& Li, K.~2006, ApJ, 641, 1087
759:
760: \bibitem[Denissenkov, Pinsonneault \& Terndrup(2006)]{dpt06}
761: Denissenkov, P.~A., Pinsonneault, M., \& Terndrup, D.~M.~2006, ApJ, 651, 438
762:
763: \bibitem[Denissenkov \& Pinsonneault(2007)]{dp07}
764: Denissenkov, P.~A., \& Pinsonneault, M.~2007, arXiv:0709.4240v1 [astro-ph]
765:
766: \bibitem[Drake et al.(2002)]{dea02}
767: Drake, N.~A., de la Reza, R., da Silva, L., \& Lambert, D.~L.~2002, AJ, 123, 2703
768:
769: \bibitem[Eggleton, Dearborn, \& Lattanzio(2006)]{eea06}
770: Eggleton P.~P., Dearborn, D.~S.~P., \& Lattanzio, J.~C.~2006, Science, 314, 1580 (EDL06)
771:
772: \bibitem[Eggleton, Dearborn, \& Lattanzio(2007)]{eea07}
773: Eggleton P.~P., Dearborn, D.~S.~P., \& Lattanzio, J.~C.~2007, in IAU Symp. 239,
774: Convection in Astrophysics, ed. F. Kupka, I. Roxburgh, \& K. Chan (Cambridge University Press), 286
775:
776: \bibitem[Gilroy \& Brown(1991)]{gb91}
777: Gilroy, K.~K., \& Brown, J.~A.~1991, ApJ, 371, 578
778:
779: \bibitem[Gratton et al.(2000)]{grea00}
780: Gratton, R.~G., Sneden, C., Carretta, E., \& Bragaglia, A.~2000, A\&A, 354, 169
781:
782: \bibitem[Grundahl et al.(2002)]{gea02}
783: Grundahl, F., Briley, M., Nissen, P.~E., \& Feltzing, S.~2002,
784: A\&A, 385, L14
785:
786: \bibitem[Herwig(2004)]{h04}
787: Herwig, F.~2004, ApJ Suppl. Ser., 155, 651
788:
789: \bibitem[Hogan(1995)]{h95}
790: Hogan, G. J.~1995, ApJ, 441, L17
791:
792: \bibitem[Johnson et al.(2005)]{jea05}
793: Johnson, C.~I., Kraft, R.~P., Pilachowski, C.~A., Sneden, C., Ivans, I.~I.,
794: \& Benman, G.~2005, PASP, 117, 1308
795:
796: \bibitem[Keller, Pilachowski, \& Sneden(2001)]{kea01}
797: Keller, L.~D., Pilachowski, C.~A., \& Sneden, C.~2001, AJ, 122, 2554
798:
799: \bibitem[Kippenhahn, Ruschenplatt, \& Thomas(1980)]{kea80}
800: Kippenhahn, R., Ruschenplatt, G., \& Thomas, H.-C.~1980, A\&A, 91, 175
801:
802: \bibitem[Lucatello et al.(2006)]{lea06}
803: Lucatello, S., Beers, T.~C., Christlieb, N., Barklem, P.~S., Rossi, S.,
804: Marsteller, B., Sivarani, T., \& Lee, Y.~S.~2006, ApJ, 652, L37
805:
806: \bibitem[Maeder(1995)]{m95}
807: Maeder, A.~1995, A\&A, 299, 84
808:
809: \bibitem[Maeder(2003)]{m03}
810: Maeder, A.~2003, A\&A, 399, 263
811:
812: \bibitem[Maeder \& Meynet(1996)]{mm96}
813: Maeder, A., \& Meynet, G.~1996, A\&A, 313, 140
814:
815: \bibitem[Masseron et al.(2006)]{mea06}
816: Masseron, T., Van Eck, S., Famaey, B., Goriely, S., Plez, B., Siess, L.,
817: Beers, T.~C., Primas, F., \& Jorissen, A.~2006, A\&A, 455, 1059
818:
819: \bibitem[Mathis, Palacios, \& Zahn(2004)]{mpz04}
820: Mathis, S., Palacios, A., \& Zahn, J.-P.~2004, A\&A, 425, 243
821:
822: \bibitem[Nollett, Busso, \& Wasserburg(2003)]{nbw03}
823: Nollett, K.-M., Busso, M., \& Wasserburg, G.~J.~2003, ApJ, 582, 1036
824:
825: \bibitem[Palacios et al.(2006)]{pea06}
826: Palacios, A., Charbonnel, C., Talon, S., \& Siess, L.~2006, A\&A, 453, 261
827:
828: \bibitem[Rood, Bania, \& Wilson(1984)]{rea84}
829: Rood, R.~T., Bania, T.~M., \& Wilson, T.~L.~1984, ApJ, 280, 629
830:
831: \bibitem[Ryan et al.(2005)]{rea05}
832: Ryan, S.~G., Aoki, W., Norris, J.~E., \& Beers, T.~C.~2005, ApJ, 635, 349
833:
834: \bibitem[Shetrone(2003)]{sh03}
835: Shetrone, M.~D.~2003, ApJ, 585, L45
836:
837: \bibitem[Smith et al.(2005a)]{sbh05}
838: Smith, G.~H., Briley, M.~M., \& Harbeck, D.~2005a, AJ, 129, 1589
839:
840: \bibitem[Smith \& Briley(2006)]{sb06}
841: Smith, G.~H., \& Briley, M.~M. 2006, PASP, 118, 740
842:
843: \bibitem[Smith \& Martell(2003)]{sm03}
844: Smith, G.~H., \& Martell, S.~L.~2003, PASP, 115, 1211
845:
846: \bibitem[Smith et al.(2005b)]{sea05}
847: Smith, V.~V., Cunha, K., Ivans, I.~I., Lattanzio, J.~C., Campbell, S., \&
848: Hinkle, K.~H.~2005b, ApJ, 633, 392
849:
850: \bibitem[Spite et al.(2006)]{spiteea06}
851: Spite, M., Cayrel, R., Hill, V., Spite, F., François, P., Plez, B., Bonifacio, P.,
852: Molaro, P., Depagne, E., Andersen, J., Barbuy, B., Beers, T. C., Nordstr\"{o}m, B., \& Primas, F.
853: 2006, A\&A, 455, 291
854:
855: \bibitem[Stancliffe et al.(2007)]{sea07}
856: Stancliffe, R.~J., Glebbeek, E., Izzard, R.~G., \& Pols, O.~R.~2007, A\&A, 464, L57
857:
858: \bibitem[Sweigart \& Mengel(1979)]{sm79}
859: Sweigart, A.~V., \& Mengel, J.~G.~1979, ApJ, 229, 624
860:
861: \bibitem[Talon \& Zahn(1997)]{tz97}
862: Talon, S., \& Zahn, J.-P.~1997, A\&A, 317, 749
863:
864: \bibitem[Talon et al.(1997)]{tea97}
865: Talon, S., Zahn, J.-P., Maeder, A., \& Meynet, G.~1997, A\&A, 322, 209
866:
867: \bibitem[Tosi(1998)]{t98}
868: Tosi, M.~1998, Space Sci. Rev., 84, 207
869:
870: \bibitem[Ulrich(1972)]{u72}
871: Ulrich, R.~K.~1972, ApJ, 172, 165
872:
873: \bibitem[Vangioni et al.(2003)]{vfea03}
874: Vangioni-Flam, E., Olive, K.~A., Fields, B.~D., \& Cass\'{e}, M.~2003,
875: ApJ, 585, 611
876:
877: \bibitem[Vauclair(2004)]{v04}
878: Vauclair, S.~2004, ApJ, 605, 874
879:
880: \bibitem[Weiss et al.(2004)]{wea04}
881: Weiss, A., Hillebrandt, W., Thomas, H.-C., \& Ritter, H.~2004, Cox \& Giuli's Principles of
882: Stellar Structure, Second Extended Edition (Cambridge Scientific Publishers: Cambridge)
883:
884: \bibitem[Weiss, Wagenhuber, \& Denissenkov(1996)]{wea96}
885: Weiss, A., Wagenhuber, J., \& Denissenkov, P.~A.~1996, A\&A, 313, 581
886:
887: \bibitem[Zahn(1992)]{z92}
888: Zahn, J.-P.~1992, A\&A, 256, 115
889:
890: \end{thebibliography}
891:
892: \clearpage
893: %-----------------------------------------------------------------
894: \pagestyle{empty}
895: \voffset=-3 cm
896: \begin{figure}
897: %\epsfxsize=15cm
898: %\epsffile [60 220 580 695] {f1.eps}
899: \plotone{f1.eps}
900: \caption{Upper panel: evolutionary tracks of a $0.83\,\msol$ model star
901: (the initial H and He mass fractions are $X=0.758$ and $Y=0.24$)
902: near the bump luminosity (solid curve --- for $\Gamma = 0.01$;
903: dashed curve --- for $\Gamma = 0.4$). Cross marks the end of
904: the first dredge-up. Bottom panel: profiles of the mean molecular
905: weight in the radiative zones of our unmixed red giant models locations of which are
906: shown with the same symbols in the upper panel. Depressions of $\mu$
907: are caused by the reaction $^3$He\,($^3$He,\,2p)$^4$He.
908: }
909: \label{fig:f1}
910: \end{figure}
911:
912: %-----------------------------------------------------------------
913: \clearpage
914: %-----------------------------------------------------------------
915: \begin{figure}
916: %\epsfxsize=15cm
917: %\epsffile [60 20 580 395] {f3.eps}
918: \plotone{f3.eps}
919: \caption{Profiles of the ratio of the thermal and molecular diffusivity (solid curve) and of
920: the quantity $K_6\equiv K/10^6$ cm$^2$\,s$^{-1}$ (dashed curve) in the radiative zone of
921: our $0.83\,M_\odot$ bump luminosity model.
922: }
923: \label{fig:f3}
924: \end{figure}
925:
926: %-----------------------------------------------------------------
927: \clearpage
928: %-----------------------------------------------------------------
929: \begin{figure}
930: %\epsfxsize=15cm
931: %\epsffile [60 20 580 395] {f6.eps}
932: \plotone{f6.eps}
933: \caption{Solutions of the dispersion relation (\ref{eq:dispersion}) as a function of
934: the horizontal turbulent diffusion coefficient. Both the only real root $\left(\omega\tau_{\rm th}\right)$
935: and the parameter $(D_{\rm h}/K)$ have been scaled appropriately to reveal the dependence
936: (\ref{eq:dulrich}) and the instability condition (\ref{eq:gmlt1}). Plotted are
937: the solutions for the fluid element diameter $d = 0.01\,H_P$ and three values of
938: $N_\mu^2 = -10^{-7}$ (this is a characteristic value for $^3$He burning in low-metallicity bump luminosity
939: stars with $M\approx 0.8\,\msol$\,; solid line), $-10^{-6}$ (short-dashed line),
940: and $-10^{-8}$ (long-dashed line).
941: }
942: \label{fig:f6}
943: \end{figure}
944:
945: %-----------------------------------------------------------------
946: \clearpage
947: %-----------------------------------------------------------------
948: \pagestyle{plaintop}
949: \voffset=0 cm
950: \begin{figure}
951: %\epsfxsize=15cm
952: %\epsffile [60 220 580 695] {f2.eps}
953: \plotone{f2.eps}
954: \caption{Comparison of the observational data from \cite{grea00}
955: for field metal-poor ($-2\la\mbox{[Fe/H]}\la -1$) low-mass stars
956: (circles) with results of our computations of the evolution of the $0.83\,\msol$ star
957: with the $^3$He-driven mixing (solid curves, equations \ref{eq:dmu})
958: and extra mixing with the rate $D_{\rm mix} = 0.02\,K$ and depth
959: $\Delta\log T = 0.19$ (dashed curves). Dotted lines show
960: predictions of the standard theory. For further details, see text.
961: }
962: \label{fig:f2}
963: \end{figure}
964:
965: %-----------------------------------------------------------------
966: \clearpage
967: %-----------------------------------------------------------------
968: \begin{figure}
969: %\epsfxsize=15cm
970: %\epsffile [60 20 580 395] {f4.eps}
971: \plotone{f4.eps}
972: \caption{Dot-dashed curve -- the empirical profile of $D_{\rm mix} = 0.02\,K$;
973: dashed curve -- a minimum threshold profile $D_{\rm inflow}=|\dot{r}|H_P$ that any $D_{\rm mix}$
974: must exceed;
975: bottom solid curve -- the $D_{\rm Kipp}$ (thermohaline convection, equation \ref{eq:dkipp})
976: profile in our unmixed bump luminosity model
977: (a hump at $r\approx 0.08\,R_{\rm sun}$ is produced by a local increase of $|\nabla_\mu|$
978: in the $^3$He-burning shell);
979: top solid curve -- $D_{\rm Kipp}$ in a model in which mixing with $D_{\rm mix}=0.02\,K$
980: has spread out the $\mu$-gradient inversion over the entire radiative zone.
981: }
982: \label{fig:f4}
983: \end{figure}
984:
985: %-----------------------------------------------------------------
986: \clearpage
987: %-----------------------------------------------------------------
988: \begin{figure}
989: %\epsfxsize=15cm
990: %\epsffile [60 20 580 395] {f5.eps}
991: \plotone{f5.eps}
992: \caption{Profiles of the mass fractions of $^3$He and CNO elements in the vicinity of
993: the H-burning shell in $0.83\,M_\odot$ bump luminosity models with
994: the heavy-element mass fractions $Z=0.002$, and $Z=0.0001$. Vertical solid
995: line segments show locations of the minimum on the $\mu$-profile. Dotted line
996: segments are placed one pressure scale height below $\mu_{\rm min}$.
997: Without overshooting, the depth of the $^3$He-driven thermohaline convection
998: would be at the locations of the solid segments. There would be no evolutionary
999: C depletion in this case, contrary to observations.
1000: }
1001: \label{fig:f5}
1002: \end{figure}
1003:
1004: %-----------------------------------------------------------------
1005:
1006: \end{document}
1007:
1008:
1009: