cond-mat0511400/ms.tex
1: %%%=======================================================
2: %%% =
3: %%% = Sample file for the class "aipproc"
4: %%% =
5: %%%=======================================================
6: 
7: \def\selectedoptions{final}
8: 
9: \documentclass[\selectedoptions]{aipproc}
10: 
11: \layoutstyle{8x11double}
12: \usepackage{fix2col}
13: %\usepackage{aipxfm}
14: \begin{document}
15: 
16: \title 
17:       [THE FREE-FREE OPACITY IN WARM, DENSE, AND WEAKLY IONIZED HELIUM]
18:       {THE FREE-FREE OPACITY IN WARM, DENSE, AND WEAKLY IONIZED HELIUM}
19: 
20: \keywords{Dense Matter, Helium, Pressure Ionization, Quantum: Molecuar Dynamics}
21: \classification{51.70.+f, 64.30.+t, 71.15.-m, 72.20.-i, 72.80.-r, 78.40.Dw}
22: 
23: \author{Piotr M. Kowalski}{
24:   address={Vanderbilt University, Nashville, TN 37235},
25:   altaddress={X-7, MS-F699, LANL, Los Alamos, NM 87545},
26:   email={kowalski@lanl.gov},
27: }
28: 
29: \iftrue
30: \author{Stephane Mazevet}{
31:   address={T-4, MS-B283, LANL, Los Alamos, NM 87545},
32:   email={smazevet@lanl.gov},
33: }
34: \author{Didier Saumon}{
35:   address={X-7, MS-F699, LANL, Los Alamos, NM 87545},
36:   email={dsaumon@lanl.gov},
37: }
38: \fi
39: 
40: \copyrightyear  {2005}
41: 
42: \begin{abstract}
43: We investigate the ionization and the opacity of warm, dense helium under conditions found in the atmospheres of cool white dwarf stars. 
44: Our particular interest is in densities up to $\rm 3 \ g/cm^{3}$ and temperatures from 1000K to 10000K. For these physical conditions various approaches
45: for modeling the ionization equilibrium predict ionization fractions that differ by orders of magnitudes. Furthermore, estimates of the density 
46: at which helium pressure-ionizes vary from $\rm 0.3$ to $\rm 14 \ g/cm^{3}$. In this context, the value of the electron-atom inverse bremsstrahlung 
47: absorption is highly uncertain. We present new results obtained from a non-ideal chemical model for
48: the ionization equilibrium, from Quantum Molecular Dynamics (QMD) simulations, and from the analysis of experimental data to better understand 
49: the ionization fraction in fluid helium in the weak ionization limit.\\
50: \end{abstract}
51: 
52: \date{\today}
53: 
54: \maketitle
55: 
56: \section{INTRODUCTION}
57: 
58: We are interested in the opacity of $\rm He$ and $\rm He/H$ mixtures for densities up to $\rm 3 \ g/cm^{3}$ and temperatures of $T\rm=1000K$ to $\rm 10000K$ 
59: to model the atmospheres of very cool white dwarf stars. Our work is motivated by the fact that there exists a wide range 
60: of predictions for the density 
61: \vskip 3pt
62: \begin{table}[h]
63: \begin{tabular}{ccc}
64: \hline
65: \tablehead{1}{c}{b}{Refference} & \tablehead{1}{c}{b}{
66: $\rho_{PI}$ ($g/cm^{3}$)} & \tablehead{1}{c}{b}{Source} 
67: \\
68: \hline
69: \cite{Ebeling} & 2 & chem \\
70: \cite{Fontaine} & 0.3 & chem \\
71: \cite{Redmer} &  4 & chem \\
72: \cite{Winisdoerffer} &  >6.5 & chem \\
73: \cite{Young} & 7.5-14 & LMTO \\
74: \cite{Fortov} & $\sim 1.5-2$ & chem \& exp \\
75: \hline
76: \end{tabular}
77: \caption{
78: Density $\rho_{PI}$ for the pressure ionization of helium from experiments (exp), chemical models (chem) and a $T\rm=0 \ K$ quantum mechanical 
79: calculation (LMTO)}
80: \end{table}
81: at which helium pressure-ionizes (Table 1).
82: This translates into large uncertainties in the number 
83: of free electrons and the value of the free-free opacity in the regime of our interest. We approach this problem by constructing more reliable models for ionization equilibrium of $\rm He$ 
84: on the basis of a chemical model, QMD simulations, and available experimental data \cite{Fortov}.
85: 
86: \section{The current state of modeling}
87: 
88: There is a large span in predictions for the density at which the pressure ionization phenomenon occurs in helium (Table 1.). 
89: 
90: For the chemical models \cite{Fortov}-\cite{Winisdoerffer}, this arises from oversimplified treatments of the physics and the presence of free parameters. 
91: The interactions between
92: atoms and ions are usually described by a polarization potential with a hard sphere cut off at short distance (e.g. \cite{Winisdoerffer}), 
93: which varies between models. Often they are simply neglected (see \cite{Redmer}). 
94: In particular, It is not correct to treat the $\rm e-He$ interaction with a polarization potential \cite{Geltman}, as we discuss below.
95: 
96: The more sophisticated quantum calculations of the $T\rm=0 \ K$ solid predict a high density for pressure ionization \cite{Young}, which we confirm 
97: with QMD simulations. On the other hand the only published experimental data on helium for the regime of pressure ionization \cite{Fortov} 
98: suggest that this phenomenon occurs at density $\rm \sim1.5-2.0 \ g/cm^{3}$.
99: 
100: \section{The electron-helium interaction}
101: 
102: The energy of a free electron in cryogenic helium has been shown experimentally to be as large as $V_{e}\rm \sim +1.5 \ eV$ (positive) at densities 
103: of $\rm \sim 0.2 \ g/cm^{3}$ \cite{Broomall}. This is in the quantitative agreement 
104: with the prediction of the Lenz potential energy \cite{Tankersley}
105: \begin{equation} V_{e}=2\pi\hbar^{2}n(\rm He \it )a/m_{e},\end{equation}
106: where $\rm a \sim 1.3 \ a.u.$ is the scattering length \cite{Tankersley}, and $n(\rm He \it )$ is the number density of helium.
107: This is because the $\rm e-He$ interaction is strongly repulsive for $\rm r>1 \ a.u.$ \cite{Kestner}, and significantly different from the polarization 
108: interaction \cite{Geltman} (Fig. 1a, solid \& dotted curves). The quantum mechanical calculations for the energy of an electron in dense helium, using the potential 
109: of \cite{Kestner}, are in good agreement with experimental data \cite{Boltjes}. This shows that free electrons interact strongly with the atoms in dense $\rm He$, but not 
110: through an attractive polarization potential. This is an important element in chemical models.
111: 
112: To introduce the interaction of quantum mechanical electrons with the atoms in a chemical model of the thermodynamics, we performed Density Functional Theory (DFT)
113: calculation of the free energy of free electron in dense $\rm He$. The results are well approximated by the Lenz formula (Fig. 1b), which we adopt in the chemical model. 
114:  
115: \begin{figure}
116: \includegraphics[width=14.8cm]{p2.ps}
117: \caption{(a) The interaction potentials for the chemical model. The lines represent the following potential curves: $\rm e-He$ of \cite{Kestner} (solid) and \cite{Geltman} (dotted),
118: $\rm He-He$ of \cite{Ross} (dashed), $\rm He-He^{+}$ of \cite{Cencek} (long dashed), and $\rm He-He_{2}^{+}$ of \cite{Scifoni} (dash-dotted). (b) The DFT free energy of a free electron in helium. 
119: The line represents the Lenz potential energy (Eq. 1).}
120: \label{fig:OldPhase} 
121: \end{figure}
122:  
123: \section{The chemical model for the ionization equilibrium}
124: 
125: We constructed a chemical model considering the following species: $\rm He,He^+,He_{2}^{+},e^{-}$.
126: The interactions between atoms and ions are described by the potentials shown in Fig. 1a. 
127: In the chemical picture, the non-ideal effects can be treated as a shift in 
128: the ionization/dissociation energies \cite{Fontaine}. Considering the reactions $\rm He \leftrightarrow He^{+}+e^{-}$ and $\rm He_{2}^{+}\leftrightarrow He+He^{+}$ the shifts are $\Delta I_{1}=\mu^{nid}(\rm He \it)-\mu^{nid}(\rm He^{+}\it)-\mu^{nid}(e)$ and 
129: $\Delta I_{2}=\mu^{nid}(\rm He_{2}^{+} \it)-\mu^{nid}(\rm He \it)-\mu^{nid}(\rm He^{+}\it)$. The non-ideal contributions to the chemical potentials for the $\rm He$ atom 
130: and the ions where obtained through the numerical solution of the Ornstein-Zernike equation in the Percus-Yevick approximation \cite{Martynow}. This approach is valid in the low ionization limit. 
131: The results are shown in Fig. 2a, 
132: where they are compared with the shift in the ionization energy extracted from conductivity data \cite{Fortov} through the relation $\sigma_{exp}/\sigma_{id}=e^{\Delta I_{1}/k_{B}T}$,
133: where $\sigma_{id}$ is the ideal gas conductivity. 
134: The agreement with the experiment is good. In this model $\rm He$ pressure-ionizes at $\rm \sim 2 \ g/cm^{3}$ because of strong attractive $\rm He-He^{+}$ interaction (Fig. 1a), 
135: which favors ionization.
136: This model is questionable, however, because this $\rm He-He^{+}$ potential inevitably leads to a bound state ($\rm He^{+}_{2}$), which would greatly reduce the interaction with
137: the other neighboring $\rm He$ atoms.
138: \begin{figure}
139: \includegraphics[width=14.8cm]{p3.ps}
140: \caption{(a) The change in the ionization/dissociation potential predicted by the chemical model. The lines represent the results for 
141: $\Delta I_{1}$ (solid, dotted) and $\Delta I_{2}$ (dashed, dash-dotted) for $T\rm=1.0 \ eV$ and $T\rm=2.0 \ eV$ respectively. The filled circles represent the value of $\Delta I_{1}^{exp}$ 
142: extracted from the experimental data of \cite{Fortov}. (b) The QMD band gap for $T\rm=0.5 \ eV$ (solid line) and $T\rm=1.5 \ eV$ (dotted line). The filled circles represent the effective value of 
143: ionization energy $\rm 24.6 \ eV -$$\Delta I_{1}^{exp}$.}
144: \label{fig:OldPhase} 
145: \end{figure}
146: 
147: \section{Quantum Molecular Dynamics results}
148: 
149: We also conducted QMD-DFT calculations of dense helium, using the Viena Ab-initio simulation package. The results in terms of band gap (ionization energy) is presented on Figure 2b. 
150: We find that in this model, helium pressure-ionizes above a density of $\rm 10 \ g/cm^{3}$. 
151: The resulting conductivities at the experimental conditions are 2 orders of magnitude smaller and show a very strong temperature dependence rather 
152: than the strong density dependence reported in \cite{Fortov}. 
153: It is well known that band gaps are underestimated using GGA functional. More accurate functionals would only increase the gap and worsen the disagreement
154: with the experimental data.
155: 
156: \begin{figure}
157: \includegraphics[width=7.0cm]{p4.ps}
158: \caption{The absorption coefficient of He obtained from QMD simulations (solid lines) for temperature $\rm T=0.5 \ eV$ and densities  
159: $\rho\rm =0.5,2.0$ and $\rm 4.0 \ g/cm^{3}$ (from the bottom to top). The dotted line represents the standard free-free absorption coefficient for $\rho \rm =0.5 \ g/cm^{3}$ 
160: \cite{Iglesias}, adopting the density of free electrons of the QMD simulation.}
161: \label{fig:OldPhase} 
162: \end{figure}
163: 
164: \section{The free-free opacity from helium}
165: 
166: The free-free (inverse bremsstrahlung) opacity of dense helium is determined by the ionization fraction and $\rm e-He$ collisions.
167: In a dense medium, the collisions can be described with the classical Drude model \cite{Jackson}. 
168: On Figure 3 we compare the QMD absorption coefficient \cite{Mazevet} with the standard free-free absorption coefficient corrected for the correlations in dense fluid but not 
169: the $\rm e-He$ collision frequency \cite{Iglesias}. For the astrophysical applications, we are interested in the spectral region with photon energies from zero to $\rm 4 \ eV$. 
170: The standard free-free frequency behaviour is erroreous, because the absorption processes for photons with small energies are driven by the frequent $\rm e-He$ collisions
171: rather than the slowly varying electric field of radiation.  
172: 
173: \section{Conclusions}
174: 
175: Our goal is to calculate the free-free absorption from dense, non-ideal, weakly ionized helium. This source of opacity depends on the ionization 
176: fraction and the frequency of $\rm e-He$ collisions. We use two different approaches to solve for the number density of free electrons in dense helium: 
177: a chemical model and QMD-DFT simulations. In the first method we have introduced a new description of the $\rm e-He$ interaction and use 
178: ab initio potentials for the He-ions interactions. This gives us a good agreement with the experiment, but the result is driven by a questionable $\rm He-He^{+}$ potential. 
179: On the other hand, the more sophisticated QMD simulations suggest that pressure ionization of $\rm He$ does not occur below $\rm 10 \ g/cm^{3}$, which is inconsistent 
180: with the conductivity data. This large discrepancy between models and the conductivity measurements is astrophysically significant. New experiments that probe the pressure
181: ionization of dense $\rm He$ are essential in finding a resolution of this problem.  
182: 
183: \begin{theacknowledgments}
184: We are grateful to Vladimir E. Fortov for providing conductivity data
185: in dense helium. This work was supported by the U.S. Department of Energy under contract W-7405-ENG-36.
186: \end{theacknowledgments}
187: %\begin{quotation}
188: \begin{thebibliography}{99}
189: \bibitem[Fortov et al, 2003]{Fortov} Fortov, V. E. et al. 2003, Journal of Experimental and Theoretical Physics, 97, 259
190: \bibitem[Ebeling et al, 1991]{Ebeling} Ebeling, W. et al 1991, Thermodynamical Properties of Hot Dense Plasmas, Physik 25, Teubner: Stuttgart
191: \bibitem[Fontaine, Graboske \& Van Horn, 1977]{Fontaine} Fontaine, G., Graboske, H. C. and Van Horn, H. M. 1977, ApJS, 35, 293
192: \bibitem[Redmer et al, 2003]{Redmer} Redmer, R., Juranek, H., Kuhlbrodt, S. and Schwarz, V. 2003, Zeitschrift fur Physikalische Chemie, 217, 783
193: \bibitem[Winisdoerffer \& Chabrier, 2005]{Winisdoerffer} Winisdoerffer, C. and Chabrier, G. 2005, Phys. Rev. E, 71, 1
194: \bibitem[Young, McMahan \& Ross, 1981]{Young} Young, D., McMahan, A. K., Ross, M. 1981, Phys. Rev B., 24, 5119
195: \bibitem[Geltman, 1972]{Geltman} Geltman, S. 1972, J. Quant. Spectrosc. Radiat. Transfer, 13, 601
196: \bibitem[Broomall, Johnson \& Onn, 1976]{Broomall} Broomall, J. R., Johnson, W. D. and Onn, D. G. 1976, Phys. Rev. B, 14, 2819
197: \bibitem[Tankersley, 1973]{Tankersley} Tankersley, L. L., 1973, Journal of Low Temperature Physics, 11, 451
198: \bibitem[Kestner et al, 1965]{Kestner} Kestner N. R., Jortner, J., Cohen, M. H. and Stuart, A. R. 1965, Phys. Rev. 140, A56
199: \bibitem[Boltjes, Graaf \& Leeuw, 1993]{Boltjes} Boltjes, B., de Graaf, C. and Leeuw, S. W. 1993, J. Chem. Phys. 98, 592 
200: \bibitem[Ross \& Young, 1986]{Ross} Ross, M., Young, d. A., 1986, Physics Letters A, 118, 463 
201: \bibitem[Cencek \& Rychlewski, 1995]{Cencek} Cencek, W. \& Rychlewski, J. 1995, J. Chem. Phys., 102, 2533
202: \bibitem[Scifoni et al, 2004]{Scifoni} Scifoni, E., Dellepiane, G. and Gianturko, F. A. 2004, The European Physical Journal D, 30,353 
203: \bibitem[Martynow, 1993]{Martynow} Martynow, G. A. 1993, Fundamental Theory of Liquids, Ch 10., Adam Hilger:Bristol.
204: \bibitem[Fritsche \& Gu, 1993]{Fritsche} Fritsche, L. and Gu, Y. M. 1993, Phys. Rev. B, 48, 7
205: \bibitem[Jackson, 1975]{Jackson} Jackson, J. D. 1975 Classical Electrodynamics, Ch 7., John Willey \& Sons: New York
206: \bibitem[Mazevet, 2004]{Mazevet} Mazevet, S., Kress, J. and Collins, L. A. in "Atomic Proceses in Plasma", AIP press, 730, 139
207: \bibitem[Iglesias, Rogers \& Saumon, 2002]{Iglesias} Iglesias, C. A., Rogers, F. J. and Saumon, D. 2002, ApJL, 569, L111
208: \end{thebibliography}
209: %\end{quotation}
210: \end{document}
211: