1: %% Start of file template.aps ****** %

2: %%

3: %%

4: %%   This file is part of the APS files in the REVTeX 4 distribution.

5: %%   Version 4.0 of REVTeX, August 2001

6: %%

7: %%

8: %%   Copyright (c) 2001 The American Physical Society.

9: %%

10: %%   See the REVTeX 4 README file for restrictions and more information.

11: %%

12: %

13: % This is a template for producing manuscripts for use with REVTEX 4.0

14: % Copy this file to another name and then work on that file.

15: % That way, you always have this original template file to use.

16: %

17: % Group addresses by affiliation; use superscriptaddress for long

18: % author lists, or if there are many overlapping affiliations.

19: % For Phys. Rev. appearance, change preprint to twocolumn.

20: % Choose pra, prb, prc, prd, pre, prl, prstab, or rmp for journal

21: %  Add 'draft' option to mark overfull boxes with black boxes

22: %  Add 'showpacs' option to make PACS codes appear

23: %  Add 'showkeys' option to make keywords appear

24: %\documentclass[aps,prl,preprint,groupedaddress]{revtex4}

25: %\documentclass[aps,prl,preprint,superscriptaddress,showpacs]{revtex4}

26: \documentclass[aps,prl,twocolumn,groupedaddress,showpacs]{revtex4}

27: 

28: % You should use BibTeX and apsrev.bst for references

29: % Choosing a journal automatically selects the correct APS

30: % BibTeX style file (bst file), so only uncomment the line

31: % below if necessary.

32: %\bibliographystyle{apsrev}

33: 

34: 

35: \newcommand{\robf}       {\mbox{\boldmath$r_0$}}

36: \newcommand{\rnbf}       {\mbox{\boldmath$r_n$}}

37: \newcommand{\fsbf}       {\mbox{\boldmath$f_s$}}

38: \newcommand{\rbf}       {\mbox{\boldmath$r$}}

39: \newcommand{\vmbf}       {\mbox{\boldmath$v_m$}}

40: \newcommand{\vsbf}       {\mbox{\boldmath$v_s$}}

41: \newcommand{\fbf}       {\mbox{\boldmath$f_0$}}

42: \newcommand{\fbfvn}       {\mbox{\boldmath$f_{vn}$}}

43: \newcommand{\ubf}       {\mbox{\boldmath$u$}}

44: \newcommand{\vbf}       {\mbox{\boldmath$v$}}

45: \newcommand{\kappabf}       {\mbox{\boldmath$\kappa$}}

46: \newcommand{\Fmfbf}       {\mbox{\boldmath$F_{mf}$}}

47: \newcommand{\nubf}       {\mbox{\boldmath$\nu$}}

48: \newcommand{\omegabf}       {\mbox{\boldmath$\omega$}}

49: \newcommand{\fdbf}       {\mbox{\boldmath$f_d$}}

50: 

51: \newcommand{\MF}{{\large{\manual META}\-{\manual FONT}}}

52: \newcommand{\manual}{rm}        % Substitute rm (Roman) font.

53: \newcommand\bs{\char '134 }     % add backslash char to \tt font

54: \def\etal{{\it et al.}}

55: \def\ie{{\it i.e.}}

56: \def\eg{{\it e.g.}}

57: \def\lap{\hbox{${_{\displaystyle<}\atop^{\displaystyle\sim}}$}}

58: \def\gap{\hbox{${_{\displaystyle>}\atop^{\displaystyle\sim}}$}}

59: 

60: \usepackage{graphicx}

61: \usepackage{subfigure}

62: 

63: \begin{document}

64: 

65: \voffset=1.cm

66: 

67: % Use the \preprint command to place your local institutional report

68: % number in the upper righthand corner of the title page in preprint mode.

69: % Multiple \preprint commands are allowed.

70: % Use the 'preprintnumbers' class option to override journal defaults

71: % to display numbers if necessary

72: %\preprint{}

73: 

74: %Title of paper

75: \title{Dynamics of Quantum Vorticity in a Random Potential} 

76: 

77: % repeat the \author .. \affiliation  etc. as needed

78: % \email, \thanks, \homepage, \altaffiliation all apply to the current

79: % author. Explanatory text should go in the []'s, actual e-mail

80: % address or url should go in the {}'s for \email and \homepage.

81: % Please use the appropriate macro foreach each type of information

82: 

83: % \affiliation command applies to all authors since the last

84: % \affiliation command. The \affiliation command should follow the

85: % other information

86: % \affiliation can be followed by \email, \homepage, \thanks as well.

87: \author{Bennett Link}

88: \affiliation{Montana State University, Department of Physics, Bozeman MT

89: 59717}

90: \email[]{link@physics.montana.edu}

91: %\homepage[]{Your web page}

92: \thanks{}

93: %\altaffiliation{}

94: 

95: %Collaboration name if desired (requires use of superscriptaddress

96: %option in \documentclass). \noaffiliation is required (may also be

97: %used with the \author command).

98: %\collaboration can be followed by \email, \homepage, \thanks as well.

99: %\collaboration{}

100: %\noaffiliation

101: 

102: \date{\today}

103: 

104: \begin{abstract}

105: I study the dynamics of a superfluid vortex in a random potential, as

106: in the inner crust of a neutron star. Below a critical flow velocity

107: of the ambient superfluid, a vortex is effectively immobilized by

108: lattice forces even in the limit of zero dissipation. Low-velocity,

109: translatory motion is not dynamically possible, a result with

110: important implications for understanding neutron star precession and

111: the dynamical properties of superfluid nuclear matter.

112: \bigskip\bigskip

113: \end{abstract}

114: 

115: % insert suggested PACS numbers in braces on next line

116: % insert suggested keywords - APS authors don't need to do this

117: %\keywords{}

118: \pacs{97.60.Jd, 26.60.-c, 47.37.+q, 97.60.Gb}

119: 

120: \bigskip

121: 

122: 

123: %\maketitle must follow title, authors, abstract, \pacs, and \keywords

124: \maketitle

125: 

126: % body of paper here - Use proper section commands

127: % References should be done using the \cite, \ref, and \label commands

128: %\section{}

129: % Put \label in7argument of \section for cross-referencing

130: %\section{\label{}}

131: %\subsection{}

132: %\subsubsection{}

133: 

134: A neutron star (NS) is expected to comprise over

135: a solar mass of distinct quantum liquids. In the inner crust (between

136: the drip density and approximately half nuclear density), a $^1_0S$

137: neutron superfluid (SF) threaded by an array of quantized vortices

138: coexists with the ionic lattice. Forces exerted by the vortex array on

139: the lattice could have important observable effects on NS spin and

140: thermal evolution (\eg,

141: \cite{shaham77,alpar_etal84,sl89,le96,swc99}), especially 

142: if vortices pin to the lattice as suggested long ago \cite{ai75}.

143: Evidence for long-period precession in some NSs (\eg, \cite{sls00}),

144: an interpretation supported by quantitative modeling \cite{alw06}, is

145: difficult to explain if vortices pin in the inner crust

146: \cite{shaham77} and presents challenges to 

147: the standard picture of the outer core

148: \cite{link03}. The critical question of whether or not pinning occurs in the

149: inner crust has been studied (\eg, \cite{jones97,jones98}) but not yet

150: satisfactorily answered. Here I show that a vortex is trapped by the

151: inner-crust lattice under very general conditions if the ambient SF

152: velocity is below a critical value.  The problem of vortex dynamics

153: and pinning in a lattice potential is also of considerable interest in

154: laboratory Bose-Einstein condensates (see,

155: \eg, \cite{bhc06}). 

156:  

157: {\em The vortex drag description.}--Most hydrodynamic studies of the

158: coupling problem begin by including a term in the SF acceleration

159: equation for the ``mutual friction'' force in a homogeneous medium.

160: For a SF of mass density $\rho_s$ flowing at velocity $\vsbf$, coupled

161: to a medium moving at velocity $\vmbf$, and neglecting the collective

162: tension of the vortex lattice, the mutual friction force density in

163: the non-rotating frame is \cite{hv56,bk61}

164: \begin{equation}

165: \Fmfbf=\beta^\prime\rho_s\omegabf\times(\vsbf-\vmbf)+

166: \beta\rho_s\nubf\times\left[\omegabf\times(\vsbf-\vmbf)\right],

167: \label{mf}

168: \end{equation}

169: where $\omegabf\equiv\nabla\times\vsbf$ is the SF vorticity, $\nubf$

170: is a unit vector in the direction of $\omegabf$, and $\beta$ and

171: $\beta^\prime$ are coefficients to be determined by a microscopic

172: calculation for the application of interest. This drag term, or

173: extensions of it to bulk neutron matter in beta equilibrium, has

174: been widely used in studies of NS hydrodynamics and precession; the

175: ``medium'' could be taken to be, for example, the lattice of the NS

176: inner crust, magnetic flux tubes of the outer core with which the vortices

177: interact, and the charged fluid of the outer core. The mutual friction

178: force of eq. [\ref{mf}] is directly related to the drag force per unit

179: length on a vortex \cite{swc99}

180: \begin{equation}

181: \fdbf=-\eta\vbf-\eta^\prime\nubf\times\vbf,

182: \label{fd}

183: \end{equation}

184: where $\vbf$ is the vortex velocity with respect to the medium, and

185: the coefficients $\eta$ and $\eta^\prime$ are related to $\beta$ and

186: $\beta^\prime$. The second term in eq. [\ref{fd}] is non-dissipative

187: and is usually assumed to be zero, as I also assume; in this case

188: $\beta=\eta_r/(1+\eta_r^2)$ and $\beta^\prime=\eta_r\beta$, where

189: $\eta_r\equiv\eta/\rho_s\kappa$ is the reduced drag coefficient and

190: $\kappa$ is the vorticity quantum $h/2m_n$ ($m_n$ is the neutron mass

191: for a neutron SF). Dissipative processes that act on a translating

192: vortex include electron scattering \citep{feibelman71,be89}, and the

193: excitation of vortex waves (Kelvin modes) through the vortex-nucleus

194: interaction

195: \citep{eb92,jones92}. All of these calculations give $\eta_r<<1$. 

196: 

197: Neglecting vortex bending and local non-dissipative forces, the motion

198: of a vortex segment follows from equating the sum of the Magnus and

199: drag forces to zero: 

200: \begin{equation}

201: \rho_s\kappabf\times\left(\vbf-\vsbf\right)

202: -\eta \vbf = 0, 

203: \label{drag}

204: \end{equation}

205: where $\kappabf$ is aligned with the vortex (the $z$ axis) and $\vsbf$

206: is now the velocity of the ambient SF in the rest frame of the medium,

207: taken to be along the $-x$ axis. In this description (henceforth, the

208: ``drag description''), the vortex segment moves at velocity

209: $\vbf\simeq\vsbf$ for low drag ($\eta_r<<1$) and $\vbf\simeq 0$ for

210: high drag ($\eta_r>>1$). The large drag limit has been taken as

211: corresponding to the pinning of vortices to defects, such as nuclei in the

212: inner crust or magnetic flux tubes in the outer core

213: \cite{swc99,sedrakian05,link06,gaj08a}. 

214: The drag description embodied in eq. [\ref{drag}], however,

215: incorrectly describes the motion of a vortex through these

216: fundamentally {\em inhomogeneous} environments. A correct description

217: of vortex motion in the inner crust requires the inclusion of two

218: additional forces: 1) the local, {\em non-dissipative} component of

219: the force exerted on the vortex by the lattice, and, 2) the elastic

220: force of the vortex. These forces qualitatively change the vortex

221: motion. Appropriate to the inner crust (but not the core), I focus

222: on a single-component SF. 

223: 

224: {\em Vortex motion through a random lattice.}--Consider the motion of

225: a vortex that is dissipatively coupled to a background lattice. The 

226: displacement vector of the vortex with respect to the $z$-axis is

227: $\ubf(z,t)=u_x(z,t)\hat{x}+u_y(z,t)\hat{y}$. The equations of motion

228: for a vortex moving in the absence of external forces can be

229: found in Sonin \cite{sonin}; a hydrodynamic description suffices

230: for excitation wavelengths significantly larger than the SF coherence

231: length $\xi$, about 10 fm in the inner crust. The force per unit

232: length exerted on the vortex by the lattice has a non-dissipative

233: contribution $\fbf$ and a dissipative contribution taken here to be

234: the drag force of eq. [\ref{fd}] with $\eta^\prime=0$, assumed to

235: hold locally, and approximated as linear in the local vortex velocity

236: (but easily generalized). The equations of motion become

237: \begin{equation}

238: T_v\frac{\partial^2\ubf}{\partial z^2}+

239: \rho_s\kappabf\times\left(\frac{\partial\ubf}{\partial t}-\vsbf\right)

240: + \fbf -\eta \frac{\partial\ubf}{\partial t} = 0, 

241: \label{eomv}

242: \end{equation}

243: where $T_v=(\rho_s\kappa^2/4\pi)\ln (\xi k)^{-1}$ is the vortex

244: self-energy (tension) for an excitation of wave number $k$. The first

245: term represents the restoring force of tension as the vortex is

246: bent. The second term is the Magnus force per unit length exerted on a vortex

247: that is moving with respect to the ambient SF. In the absence of

248: external forces, the vortex would remain straight and motionless with

249: respect to the solid. An essential feature of vortex dynamics is that

250: the local vortex velocity is determined entirely by external forces

251: and by the shape of the vortex, and not by inertial forces (for

252: temperatures small compared to the pairing gap). In

253: the absence of external forces, the solutions to eqs. [\ref{eomv}] are

254: circularly-polarized, diffusive waves (Kelvin waves) with frequencies

255: $\omega_k = \pm T_v k^2/\rho_s\kappa$ \cite{kelvin}. Characteristic

256: bending scales of the vortex in the problem are $\sim 500$ fm (see

257: below), and fixing $\ln (\xi k)^{-1}=3$ is an adequate approximation. I

258: ignore quantum effects on the vortex motion and thermal excitations.

259: 

260: Nuclei in the solid with which the superfluid coexists exert local 

261: forces on the vortex which vary over a length scale of order the

262: average nuclear spacing $a$, typically 30-50 fm in the inner

263: crust. Calculations of the vortex-nucleus interaction energy $E_{vn}$

264: give values of up to $\sim 1$ MeV, attractive

265: in some regions and repulsive in others

266: \citep{dp06,abbv07}; the corresponding characteristic force on a vortex

267: segment of length $a$ is $F_m\equiv a^{-1}E_{vn}$.  

268: The solid is most likely amorphous

269: \cite{jones01}, in which case the vortex interacts with an 

270: effectively {\em random potential} 

271: Even if the lattice does possess long-range crystalline

272: order, vortices will be aligned with the SF rotation axis on average

273: which will not, in general, coincide with any symmetry axis of the

274: solid; a random lattice approximation should be adequate also in this

275: situation. In the denser regions of the NS inner crust, $T_v$ is of

276: order an MeV fm$^{-1}$. Defining a dimensionless strength parameter

277: $s\equiv F_m/T_v$ ($>0$ for an attractive potential), estimates of

278: $F_m$ from Refs. \cite{dp06,abbv07} give $10^{-2}\le\vert s\vert\le

279: 0.1$. A vortex is thus a stiff object; the vortex-nucleus interaction

280: can bend a vortex by an amount $a$ only over a length scale of order

281: $\gap 10a$

282: \cite{lc02}; a vortex segment of length $a$ can be regarded as

283: straight to a good approximation and the lattice force on the vortex

284: can be taken as acting in the $x-y$ plane at any given $z$ along the

285: vortex. For the force exerted by a single nucleus at the origin on a

286: vortex segment of length $a$, I take a parameterized central force in

287: the $x-y$ plane:

288: \begin{equation}

289: \fbfvn(\ubf)=-1.7F_m\frac{\ubf}{r_p}{\rm e}^{-u^2/2r_p^2} 

290: \label{force}

291: \end{equation}

292: The force has a maximum magnitude of $F_m$ at $u=r_p$, where $r_p$ is

293: the effective range of the potential, henceforth taken to be

294: $r_p=a/2$.  Dividing the vortex into segments of length $a$, let the

295: total non-dissipative force on the $i$-th segment be

296: \begin{equation}

297: \fbf(z_i) =  \sum_j \fbfvn(\ubf_i-\rbf_j),

298: \label{fl}

299: \end{equation}

300: where the nuclei are randomly placed at locations $\rbf_j$ in planes

301: separated by $a$ and parallel to the $x-y$ plane. For simplicity, each

302: summation is only over nuclei in the $i$-th plane.  In the following,

303: lengths will be expressed in units of $a$, time in units of $t_D\equiv

304: \rho_s\kappa a^2/T_v$ (the characteristic diffusion time of a Kelvin

305: wave of wavenumber $k=a^{-1}$ along the segment) and velocities in

306: units of $a/t_D$. As described below, the sign of $s$ is unimportant

307: for $r_p\sim a$; I choose $s>0$, corresponding to attractive nuclear

308: potentials, for illustration. I focus on the regime

309: $0.1\ge\eta_r\ge 0$ which widely brackets the range found by all drag

310: calculations, and take $s=0.1$. 

311: 

312: \begin{figure}[t]

313: \centering

314: \subfigure

315: {

316:     \includegraphics[width=.45\linewidth]{f1.pdf}

317: }

318: \subfigure

319: {

320:      \includegraphics[width=.47\linewidth]{f2.pdf}

321: }

322: \caption{{\em Left panel:} vortex position, averaged over its

323: length, in the presence of an external SF flow field of velocity

324: $v_s=10^{-3}$ and no dissipation over $5\times 10^3$ time

325: units. Without the random potential, the vortex would translate to the

326: left with the SF by five units. The non-dissipative lattice force

327: effectively pins the vortex, confining it to a meandering trajectory

328: over a length scale of order $a$. {\em Right panel:} Length-averaged

329: vortex speed in response to local SF flow at

330: $v_s$. Velocities have been scaled by $s^{-3/2}$. For relatively high

331: $v_s$, the vortex translates at a speed $\sim v_s$, but is trapped for

332: $v_s<v_c\simeq s^{3/2}$, independent of drag.}

333: \label{motion}

334: \end{figure}

335: 

336: \begin{figure*}[t]

337: \centering

338: \includegraphics[width=.285\linewidth]{f3.pdf} 

339: \includegraphics[width=.3\linewidth]{f4.pdf} 

340: \includegraphics[width=.31\linewidth]{f5.pdf}

341: \caption{Motion of an example vortex segment through one plane of the

342: lattice for $\eta_r=0.1$. {\em Left panel:} For $v_s>v_c\sim s^{3/2}$, the

343: vortex translates with the superfluid, with significant changes in

344: direction as it interacts with nuclei (denoted by circles). For $v_s<v_c$, the

345: segment damps to a pinned configuration that brings it closer to

346: nuclei. {\em Middle panel:} Comparison of the damped motion for

347: $v_s=0$ and $v_s<v_c$, showing that finite $v_s$ has little effect on

348: the motion in this regime. {\em Right panel:} Detail of the box

349: region of the middle figure. Circles on the lines indicate points of

350: equal time. The result of finite $v_s$ is to displace the pinned

351: vortex slightly in the $y$ direction.  }

352: \end{figure*}

353: 

354: 

355: Comparison of terms in eq. [\ref{eomv}] shows that the dynamics are

356: essentially determined by the non-dissipative lattice force and $s$ in

357: the regime $v_s<v_c$. 

358: To determine the character of the dynamical regimes of vortex motion

359: and the scaling relations that define them,

360: eqs. [\ref{eomv}]-[\ref{fl}] were solved numerically. The vortex was 

361: divided into 50 zones of length $a$, with periodic boundary conditions

362: applied at the ends, in a random lattice consisting of $10^3$ nuclei

363: per zone. Fig. \ref{motion} (left panel) shows that the vortex is

364: trapped by the random potential for finite $v_s$ even for zero

365: drag. The chief result of the numerical analysis is shown in the right

366: panel of Fig. 1. Trapping occurs for $v_s$ below a critical velocity

367: $v_c\simeq s^{3/2}$, {\em independent of drag} in the regime

368: $\eta_r<<1$. For $v_s>v_c$, the vortex moves at $\vbf\simeq\vsbf$; in

369: this regime, the drag description of eq. [\ref{drag}] is

370: adequate. Below $v_c$, however, the vortex velocity is effectively

371: zero. The drag description fails to show this transition to the pinned

372: state because it excludes both the non-dissipative lattice force and the

373: finite vortex tension. The value of the critical velocity $v_c\sim

374: s^{3/2}$ follows from consideration of a vortex in static equilibrium,

375: as discussed elsewhere \cite{jones98,lc02}. Only if the vortex had

376: infinite tension, one of the assumptions implicit in the drag description and

377: eq. [\ref{drag}], would translatory states exist for any $v_s$.

378: 

379: Dissipative processes will damp the vortex to a stationary, pinned

380: configuration.  Fig. 2 shows an example segment of the long vortex

381: moving under drag. For $v_s<v_c$, the trajectory is essentially the

382: same as for $v_s=0$ since the non-dissipative force dominates the

383: global Magnus force in this velocity regime. Finite $v_s$ (below $v_c$),

384: produces only a small displacement of the vortex segment's final,

385: damped position. Inspection of eq. [\ref{eomv}] suggests that a

386: dragged vortex will damp to a pinned position over a characteristic

387: time $\sim (s\eta_r)^{-1}$, and numerical experiments confirm

388: this. The damping time is $\sim 100$ for the examples shown in

389: Fig. 2. Fig. 3 shows the initial motion and damping of the vortex in

390: the $x-z$ plane. The initially-straight vortex forms kinks immediately

391: in the presence of the random potential, and by $t=10$ the vortex has

392: been excited to amplitudes of order $a$, the length scale of the

393: random potential. By $t=10^3$, the vortex has damped to a stationary

394: pinned configuration with bends over a characteristic length scale of

395: $\sim 10a$. Because the vortex has large tension, it cannot bend to

396: intersect every nucleus, but assumes a shape that strikes the best

397: compromise between the energy gain of being close to a nucleus and the

398: cost of bending the vortex.

399: 

400: \begin{figure}[b]

401: \centering

402: \includegraphics[width=8.cm]{f6.pdf} 

403: \caption{Motion and pinning of the vortex in the $x-z$ plane for the

404: simulation of Fig. 2 with $\vsbf=0$.}

405: %\vspace*{1.cm}}

406: \end{figure}

407: 

408: 

409: Further simulations show that these conclusions are unaffected if the

410: attractive nuclear potentials are replaced by repulsive ones (though

411: $v_c$ is slightly reduced for the force of eq. [\ref{force}]). This

412: result is not surprising, since for $r_p\sim a$ the lattice potential

413: effectively turns inside out. A vortex segment now damps to a position

414: that maximizes its distance from the nearest nuclei to the extent

415: possible against opposing tension forces. For $\eta_r=0$ the vortex is

416: trapped by the lattice as for the case of attractive nuclei. Since

417: these results are independent of the value of $\eta_r$ provided

418: $\eta_r<<1$, they are also independent of the assumption that $\eta_r$

419: is a constant.

420: 

421: Recent calculations of $E_{vn}$, using the local density approximation

422: \cite{dp06} and

423: mean field theory \cite{abbv07} are not in complete agreement. Though

424: both calculations predict comparable interaction energies ($\sim 1$

425: MeV) above an inner-crust density of $\sim 10^{13}$ g cm$^{-3}$, they

426: do not agree on the densities where the interaction is the strongest.

427: The inevitability of pinning is nevertheless a robust result; it is

428: insensitive to the exact density range in which $E_{vn}$ is strongest,

429: the length scale of the potential, or even the sign of the

430: potential. Within these uncertainties, the critical velocity above

431: which vortices will translate is

432: \begin{equation}

433: v_c\sim

434: s^{1/2}\frac{F_m}{\rho_s\kappa a} =10^6-10^7 \mbox{ cm s$^{-1}$}.

435: \label{vc}

436: \end{equation}

437: The lattice

438: will trap the vortex for much smaller values of $E_{vn}$ as well,

439: though $v_c$ will be lowered according to eq. [\ref{vc}]. 

440: 

441: {\em Conclusions and implications.}---Pinning of vortices to the

442: lattice of the NS inner crust below a critical value of $v_s$ appears

443: to be inevitable, whether the interaction is attractive or repulsive

444: (and provided $E_{vn}$ exceeds the stellar temperature, typically

445: $\sim 0.01$ MeV). For interactions of order $\sim 1$ MeV per nucleus,

446: the pinned vortex lattice becomes unstable for $v_s>v_c\sim 10^6-10^7$

447: cm s$^{-1}$, and vortices will translate approximately with the SF if

448: the drag is low. For critical velocities this large, the pinned

449: superfluid can store enough angular momentum to drive the giant

450: glitches seen in pulsars \cite{lc02}.  If pinning in regions of the

451: inner crust does occur, however, interpretation of the putative

452: precession seen in some pulsars becomes problematic; as noted by

453: Shaham

454: \cite{shaham77}, if even a small portion of the inner crust vorticity

455: is tightly coupled to the solid, the star will precess much faster

456: than the period of $\sim 1$ yr indicated by observations \cite{sls00}.  In the

457: outer core, where the protons are predicted to form a type-II

458: superconductor, vortices are expected to pin against a disorganized

459: system of flux tubes and a similar difficulty arises in explaining

460: long-period precession \cite{link03}. Anywhere there is pinning,

461: however, vortices can creep ($v<<v_s$) through thermal activation or

462: quantum tunneling processes \cite{alpar_etal84,leb93} not considered

463: here. Vortex creep is incompatible with long-period precession if it

464: is a high-drag process \cite{swc99,link06}.

465: 

466: \enlargethispage*{10pt}

467: 

468: The core neutron-charge mixture could be unstable to the formation of

469: SF turbulence in a precessing star \cite{peralta_etal05,gaj08a},

470: though magnetic stresses might suppress such an instability

471: \cite{vl08}. [Turbulent instabilities might not occur in the 

472: single-component SF of the inner crust]. These analyses used the drag

473: description of vortex motion which does not correctly describe vortex

474: motion through a potential; it is necessary to include non-dissipative

475: pinning forces. The mutual friction force of eq. [\ref{mf}] should be

476: replaced with the {\em total} force per unit volume exerted on the SF

477: by the vortex array. For a single-component SF, this force is (see,

478: \eg, \cite{swc99}):

479: \begin{equation}

480: {\mathbf F}=\rho_s \left\{\vbf(\vsbf) - \vsbf\right\}\times\omegabf,

481: \label{ftot}

482: \end{equation}

483: where $\vbf(\vsbf)$ is local vortex velocity with respect to the

484: background, averaged over many vortices, in response to a local flow

485: $\vsbf$. In the approximations made here, $\vbf=0$ for $v_s<v_c$, and

486: ${\mathbf F}$ reduces to the Magnus force per unit volume on pinned

487: vortices. At finite $\vsbf$, however, vortices can move slowly through

488: vortex creep, a process which dissipates energy at a rate

489: $\rho_s\vbf\cdot (\vsbf\times\omegabf)$ per unit volume; the

490: dissipative force in the fluid would be small if $\vbf(\vsbf)$ is

491: nearly orthogonal to $\vsbf\times\omegabf$, while the non-dissipative

492: force would be, in any case, nearly equal to the Magnus force on

493: perfectly pinned vortices. It would be interesting to study NS star

494: modes under the assumption that creep proceeds with little

495: dissipation, the opposite limit of that studied so far.

496: 

497: % If you have acknowledgments, this puts in the proper section head.

498: \begin{acknowledgments}

499: I thank I. Wasserman and C. M. Riedel for useful discussions. 

500: \end{acknowledgments}

501: 

502: % Create the reference section using BibTeX:

503: 

504: \bibliography{references}

505: 

506: \end{document}

507: