1: 

2:  \documentclass[prd, twocolumn, showpacs, showkeys, floatfix]{revtex4}

3:  % a4paper, nofootinbib, 

4: 

5: % Packages:

6: 

7: \usepackage{latexsym} % Gets \Box etc

8: \usepackage{amssymb}  % \gtrsim, \geqslant, etc etc: see amsguide.ps

9: \usepackage{amsfonts} % \mathfrak and \mathbb{x} (Blackboard bold)

10: \usepackage{graphicx}       % For PostScript figures

11: %\usepackage{multirow}  % for multirows in tables

12: \usepackage[dvips]{color}

13: 

14: %\addtolength{\topmargin}{-1cm} % Raise text on page for archive

15: 

16: %============================================================

17: %  Generic Abbreviations:

18: %============================================================

19: 

20: \newcommand{\al}{\alpha}

21: \newcommand{\be}{\beta}

22: \newcommand{\ga}{\gamma}

23: \newcommand{\Ga}{\Gamma}

24: \newcommand{\de}{\delta}

25: \newcommand{\De}{\Delta}

26: \newcommand{\ep}{\varepsilon}

27: \newcommand{\eps}{\epsilon}

28: \newcommand{\ze}{\zeta}

29: \newcommand{\ka}{\kappa}

30: \newcommand{\la}{\lambda}

31: \newcommand{\La}{\Lambda}

32: \newcommand{\ph}{\varphi}

33: \newcommand{\si}{\sigma}

34: \newcommand{\Si}{\Sigma}

35: \renewcommand{\th}{\theta}   % LaTeX: \th already defined

36: \newcommand{\Up}{\Upsilon}

37: \newcommand{\om}{\omega}

38: \newcommand{\Om}{\Omega}

39: 

40: %

41: \newcommand{\imp}{~\Rightarrow}

42: %\newcommand{\to}{\rightarrow}

43: \newcommand{\p}{\partial}

44: \newcommand{\<}{\langle} 

45: \renewcommand{\>}{\rangle} % LaTeX: \> already defined

46: \newcommand{\ul}{\underline}

47: \newcommand{\txt}{\textstyle}

48: \newcommand{\sst}{\scriptstyle}

49: \newcommand{\dsp}{\displaystyle}

50: \newcommand{\ns}{\normalsize}

51: \newcommand{\ad}{\dagger}

52: \newcommand\eqn[1]{(\ref{#1})}      % parentheses around the LaTex "ref" macro

53: \newcommand\Eqn[1]{Eq.~(\ref{#1})}  % includes ``Eq.'' in front

54: \newcommand{\e}{{\rm e}}

55: \newcommand{\beq}{\begin{equation}}

56: \newcommand{\eeq}{\end{equation}}

57: \newcommand{\ba}{\begin{array}}

58: \newcommand{\bea}{\begin{eqnarray}}

59: \newcommand{\ea}{\end{array}}

60: \newcommand{\eea}{\end{eqnarray}}

61: \newcommand{\bc}{\begin{center}}

62: \newcommand{\ec}{\end{center}}

63: \newcommand{\ben}{\begin{enumerate}}

64: \newcommand{\een}{\end{enumerate}}

65: 

66: \newcommand{\dslash}{{\partial\kern-0.55em/}}

67: %\newcommand{\Dslash}{{D\kern-0.65em/}}

68: 

69: \def \muslash {\mu \kern-0.5em\slash}

70: 

71: % Common math/phys abbreviations

72: \renewcommand{\O}{{\cal O}}

73: \newcommand{\order}{{\cal O}}

74: \newcommand{\tr}{\hbox{tr}}

75: \newcommand{\Tr}{\hbox{Tr}}

76: \newcommand{\hc}{{\rm h.c.}}

77: \renewcommand{\Re}{{\rm Re}\,}

78: \newcommand{\ie}{{i.e.}}

79: \newcommand{\eg}{{e.g.}}

80: %\def\sc{\scriptstyle}

81: %\def\scsc{\scriptscriptstyle}

82: \newcommand{\seq}{\!=\!}        % for eqs in text

83: \newcommand{\Dash}{\boldmath $-$}  % for item symbol

84: \newcommand{\skipover}[1]{}

85: \newcommand{\nn}{\nonumber \\}

86: \newcommand{\half} {{\txt \frac{1}{2}}}

87: \newcommand{\third}{{\txt \frac{1}{3}}}

88: \newcommand{\quarter}{{\txt \frac{1}{4}}}

89: \newcommand{\twothirds}{{\txt \frac{2}{3}}}

90: \newcommand{\sixth}{{\txt \frac{1}{6}}}

91: \newcommand{\eighth}{{\txt \frac{1}{8}}}

92: 

93: % 1/3 neg thin space, for fine-tuning formulae

94: \def\back{\mskip-.333\thinmuskip}

95: % Phantom minus sign: very useful for lining things up!

96: \def\phm{\phantom{-}}

97: 

98: 

99: 

100: % ======================================================================

101: % Abbreviations specific to this paper:

102: % ======================================================================

103: 

104: \newcommand{\eV}{{\rm eV}} 

105: \newcommand{\keV}{{\rm keV}} 

106: \newcommand{\MeV}{{\rm MeV}} 

107: \newcommand{\GeV}{{\rm GeV}} 

108: \newcommand{\vp}{{\mathbf p}}    % vector p

109: \newcommand{\vk}{{\mathbf k}}    % vector k

110: 

111: \newcommand{\pslash}{{p\kern-0.45em/}}

112: \newcommand{\kslash}{{k\kern-0.5em/}}

113: 

114: \newcommand{\three}{\ensuremath{{\bf 3}}}

115: \newcommand{\thrS}{\ensuremath{{\bf 3}_S}}

116: \newcommand{\thrbar}{\ensuremath{\bar{\bf 3}}}

117: \newcommand{\thrbarA}{\ensuremath{\bar{\bf 3}_A}}

118: \newcommand{\six}{\ensuremath{{\bf 6}}}

119: \newcommand{\sixS}{\ensuremath{{\bf 6}_S}}

120: \newcommand{\one}{\ensuremath{{\bf 1}}}

121: \newcommand{\oneA}{\ensuremath{{\bf 1}_A}}

122: 

123: \newcommand{\eight}{${\bf 8}$}

124: \newcommand{\psibar}{{\bar\psi}}

125: 

126: \newcommand{\Lag}{{\cal L}}

127: \newcommand{\onesc}{\ensuremath{({\rm 1SC})^3}}

128: \newcommand{\csl}{\ensuremath{({\rm CSL})^3}}

129: \newcommand{\twocsl}{2SC+CSLs}

130: \newcommand{\twoonesc}{2SC+1SCs}

131: 

132: % Used in appendix:

133: % \newcommand{\D}{{\cal D}}

134: % \newcommand{\xbar}{\bar{x}}

135: % \newcommand{\pbar}{\bar{p}}

136: % \newcommand{\ftfermi}[1]{\sqrt{V}\sum_{n}\int\frac{d^3p}{(2\pi)^3}\,e^{#1 \pbar\cdot\xbar}}

137: % \newcommand{\ftbose}[1]{T\sum_{n}\int\frac{d^3p}{(2\pi)^3}\,e^{#1 \pbar\cdot\xbar}}

138: % 

139: % \newcommand{\config}{\int_0^{1/T} d\tau\int_V d^3x\;}

140: % \newcommand{\momentum}{V\sum_n\int\frac{d^3p}{(2\pi)^3}\;}

141: % \newcommand{\sumX}{\sum_X}

142: % \newcommand{\sumP}{\sum_P}

143: % \newcommand{\intdp}{\int\frac{d^3p}{(2\pi)^3}}

144: 

145: % Marking old and new text:

146: \newcommand{\com}[1]{{\sf\color[rgb]{0,0,0.8}[#1]}}

147: \newcommand{\new}[1]{{\sf\color[rgb]{0.8,0,0}#1}}

148: %\newcommand{\new}[1]{#1}

149: \usepackage[normalem]{ulem}  % \sout{old text} for strikeout

150: \newcommand{\old}[1]{{\sf\color[rgb]{0.7,0.7,0.7}\sout{#1}}}

151: 

152: 

153: % ************************************************************************

154: % **************************** DOCUMENT BEGINS ***************************

155: % ************************************************************************

156: 

157: 

158: \begin{document}

159: 

160: \title{\bf 

161: Single-flavour and two-flavour pairing in 

162: three-flavour quark matter

163: }

164: 

165: \author{Mark G. Alford}

166: %\email{alford@wuphys.wustl.edu}

167: \affiliation{Department of Physics, Washington University\\ 

168:   St Louis, MO, 63130, USA}

169: \author{Greig A. Cowan}

170: %\email{g.cowan@physics.gla.ac.uk}

171: \affiliation{Department of Physics and Astronomy, University of Glasgow\\

172:   Glasgow, G12 8QQ, UK}

173: 

174: %%\preprint  % something wrong with this in ReVTeX4

175: %%\newcommand{\preprintno}{

176: %%  \normalsize GUTPA??

177: %%}

178: 

179: \date{2005-Dec-8}

180: 

181: % \begin{titlepage}

182: % \maketitle

183: % \def\thepage{}          % No page number on title page

184: 

185: \begin{abstract}

186: %       1         2         3         4         5         6

187: We study single-flavour quark pairing (``self-pairing'')

188: in colour-superconducting phases of quark matter,

189: paying particular attention to the difference between

190: scenarios where all three flavours undergo single-flavour pairing,

191: and scenarios where two flavours pair with each other

192: (``2SC'' pairing) and the remaining flavour self-pairs.

193: We perform our calculations in the mean field approximation

194: using a pointlike four-fermion interaction based on 

195: single gluon exchange.

196: We confirm the result from previous weakly-coupled-QCD 

197: calculations, that when all three flavours self-pair

198: the favored channel for each is colour-spin-locked (CSL) 

199: pseudoisotropic pairing.

200: However, we find that when the up and down quarks undergo 2SC pairing, 

201: they induce a colour chemical potential that disfavors the CSL phase.

202: The strange quarks then self-pair in a ``polar'' channel that

203: breaks rotational invariance, although the CSL phase may survive

204: in a narrow range of densities.

205: \end{abstract}

206: %%\end{titlepage}  % taken out to comply with ReVTeX4

207: 

208: \pacs{12.38.-t,25.75.Nq}

209: \keywords{quark matter; colour superconductivity; 

210: 2SC; CSL; neutrality; free energy}

211: \preprint{GUTPA ???}

212: 

213: \maketitle

214: 

215: \renewcommand{\thepage}{\arabic{page}}

216: %\setcounter{page}{1}

217: 

218: %--------------------------------------- 

219:                         % Body of paper begins

220: \section{INTRODUCTION}

221: \label{sec:intro}

222: 

223: 

224: 

225: Matter at high density and sufficiently low temperature is expected to

226: form a colour superconducting condensate of quark pairs

227: \cite{BarroisPhD,Bailin:1983bm,ARW2,Rapp:1997zu} (for reviews, see

228: Ref.~\cite{Rajagopal:2000wf,alford_review,Alford:2002ng,Nardulli:2002ma,Schafer:2003vz}).

229: Such a phase of matter may exist in the cores of

230: compact stars \cite{Glendenning} or be created during low energy

231: heavy-ion collisions \cite{Senger:2005pu}. At high density, quarks will fill

232: up the available energy states to form a Fermi surface. Since 

233: two quarks in the antisymmetric colour antitriplet channel

234: experience an attractive interaction, we expect a

235: BCS pairing instability \cite{bcs1}

236: at the Fermi surface. The instability is

237: resolved via the opening of a energy gap in the quasiparticle

238: spectrum. The quark pairing breaks the $SU(3)_{\text{colour}}$ gauge

239: symmetry of QCD, justifying the name colour superconductor.

240: 

241: The strong interaction is most attractive between quarks in

242: the colour-antisymmetric spin-0 channel, but Fermi-Dirac statistics 

243: then require flavour-antisymmetric pairing, involving two different

244: flavours. Well-studied candidates include the 2SC

245: \cite{ARW2,Rapp:1997zu,Buballa:2002xv} and CFL

246: \cite{ARW3,Alford:1999pa,Buballa:2001gj} phases of quark matter.

247: At asymptotically high

248: density, the favored phase of quark matter is the CFL phase.

249: 

250: However, the cores of compact stars are not asymptotically dense and

251: it becomes necessary to consider real-world effects such as a

252: non-zero strange quark mass, electrical and colour charge neutrality

253: and $\be$-equilibrium. Taken together, these constraints act to

254: separate the Fermi surfaces of the different quark flavours.

255: This means that as density decreases from the asymptotic regime

256: it becomes harder and harder to maintain the CFL condensate,

257: and we expect transitions to other phases.

258: In such a context it is natural to look for single-flavour pairing patterns,

259: and many different ones have been found

260: \cite{IwaIwa,TS1flav,BHO,Alford:2002kj,Schmitt:2004hg}. 

261: Calculations in weak-coupling QCD

262: indicate that for a single isolated flavour

263: the colour-spin-locked (CSL) phase has the

264: lowest free energy. However, in

265: the realistic context of quark matter in neutron

266: stars there are three flavours present. In this paper

267: we explore the single-flavour phases in such a context,

268: using a Nambu-Jona-Lasinio model. We find that if

269: the up and down quarks undergo two-flavour (2SC) pairing

270: then this typically induces a non-zero colour chemical potential

271: that disfavors CSL pairing for the remaining flavour,

272: which then self-pairs in a different ``transverse polar'' channel instead,

273: which we will simply label as ``1SC''.

274: 

275: The 1SC/polar channel was studied in the NJL context 

276: in Ref.~\cite{Alford:2002kj}, where we surveyed the pairing patterns 

277: that factorize into a product of colour, flavour, and Dirac (spin)

278: structures (this ansatz excludes the CSL phase, where colour

279: and spin are intertwined).

280: We used an NJL model with various plausible interactions,

281: and found that for a single flavour the most attractive channel

282: was the (\thrbarA,\thrS,1,$+$)($C\ga_3$) channel, where the

283: notation translates as (colour antisymmetric $\thrbar$,

284: flavour symmetric $\six$, spin 1, parity even), with 

285: Dirac structure $C\ga_3$. Pairing between quarks of the same

286: flavour in this channel was found to have a gap parameter of $\sim

287: 1-10$ MeV. Unlike the CSL phase, this condensate breaks rotational

288: symmetry.

289: In weak-coupling QCD calculations this is called the

290: ``transverse polar'' phase \cite{Schmitt:2004et}.

291: 

292: \begin{figure}[t]

293:  \includegraphics[width=0.5\textwidth]{./figures/1sc_levels.eps}

294:  \caption[Pictorial representation of 1SC pairing in neutral quark matter.]{

295: Pictorial representation of simple single-flavour pairing in

296: neutral quark matter. This will be referred to as the \onesc\ phase in

297: the text. The requirement of electric neutrality and a nonzero strange quark

298: mass forces the Fermi momenta of the three flavours apart.

299: Two colours of each flavour form (\thrbarA,\thrS,1,$+$)($C\ga_3$)

300: Cooper pairs (1SC$u$, 1SC$d$ and 1SC$s$). The third colour of each

301: flavour forms (\sixS,\thrS,0,$+$)($C\ga_0\ga_5$) pairs.

302: }

303: \label{fig:1sc_levels}

304: \end{figure}

305: 

306: \begin{figure}[t]

307: \includegraphics[width=0.5\textwidth]{./figures/2sc_levels.eps}

308: \caption[Pictorial representation of 2SC and 1SC pairing 

309: in neutral quark matter.]{Pictorial representation of 

310: simultaneous 2SC and 1SC

311: pairing in neutral quark matter. This will be referred to as the

312: \twoonesc\ phase in the text. This will only occur if the

313: condensation energy of the 2SC pairing is strong enough to offset

314: the cost of dragging the red and green $u$ and $d$ Fermi momenta

315: away from the values dictated by electrical neutrality and the

316: strange quark mass, to a common value.

317: \label{fig:2sc_levels}}

318: \end{figure}

319: 

320: \begin{figure}[t]

321: \includegraphics[width=0.25\textwidth]{./figures/CSL_levels.eps}

322: \caption[Pictorial representation of CSL pairing

323: in neutral quark matter.]{Pictorial representation of CSL pairing in

324:   neutral quark matter. This will be referred to as the \csl\ phase in

325:   the text. It is composed of up, down and strange quark CSL

326:   condensates, labelled CSL$u$, CSL$d$ and CSL$s$ respectively. The

327:   requirement of electric neutrality and a nonzero strange quark mass

328:   forces the Fermi momenta of the three flavours apart. The red, green

329:   and blue colours of each flavour pair in a colour-antisymmetric

330:   channel.

331:   \label{fig:CSL_levels}}

332: \end{figure}

333: 

334: \begin{figure}[t]

335: \includegraphics[width=0.5\textwidth]{./figures/2sc_CSL_levels.eps}

336: \caption[Pictorial representation of 2SC and CSL pairing 

337: in neutral quark matter.]{Pictorial representation of 2SC and CSL

338: pairing in neutral quark matter. This will be referred to as the

339: \twocsl\ phase in the text. Colour neutrality will create a small

340: $\O(\De^2/\mu)$ splitting between the Fermi momenta of the blue $s$

341: quarks and $s$ quarks of the other two colours. If this splitting is

342: sufficiently large it will prevent the CSL$s$ condensate from forming.

343: \label{fig:2SC_CSL_levels}}

344: \end{figure}

345: 

346: In this paper we do not attempt a comprehensive study of all the

347: phases that have been suggested for quark matter. Our purpose is to

348: show that the presence of 2SC pairing can affect the single-flavour

349: pairing of the remaining flavour, so we restrict our study to

350: polar and CSL pairing of individual flavours, and 2SC pairing

351: of the up and down quarks. We do not take into account the

352: CFL phase, nor its gapless variant gCFL \cite{Alford:2003fq,Alford:2004hz}, 

353: nor crystalline (LOFF) pairing \cite{OurLOFF,Bowers:2002xr,Bowers_thesis},

354: and so on. 

355: Our study is relevant to regions of the phase diagram where

356: three flavours of quark are present, but there is either

357: no cross-flavour pairing, or only pairing between the

358: up and down quarks (``2SC+s'' in the the nomenclature of

359: Ref.~\cite{Alford:1999pa}). Such regions have been found to exist

360: in studies of the QCD phase diagram using NJL models.

361: For example Ref.~\cite{Ruster:2005jc} finds a 2SC phase with strange quarks

362: for stronger diquark coupling ($G_D=G_S$)

363: at $T=0$, over the chemical potential range 

364: $398~\MeV \lesssim \mu \lesssim 412~\MeV$

365: (see their Figs.~2 and 6). Ref.~\cite{Abuki:2005ms} also

366: finds such a phase (see their Fig.~7). It is also interesting to

367: note that there is some chance our considerations might be relevant

368: to the early life of the star, when $T\gtrsim 1~\MeV$,

369: neutrinos are trapped, and the resultant lepton number chemical

370: potential can favor a 2SC phase with strange quarks \cite{Steiner:2002gx}.

371: However some studies only find up and down quarks at this time 

372: \cite{Ruster:2005ib}. Moreover,

373: single flavour phases will only be relevant if they

374: have large enough gaps to survive at these temperatures. We find that their

375: critical temperatures are of order 1~\MeV (see Fig.~\ref{fig:1SC_CSL_gaps_T}),

376: but this result is sensitive to the cutoff 

377: (see end of Section \ref{sec:results}) so it remains an open

378: question whether they can survive during the era of neutrino trapping.

379: 

380: 

381: The paper has the following structure. Section \ref{sec:1SC/CSL}

382: summarizes the 1SC and CSL pairing patterns.

383: Sections \ref{sec:outline} and \ref{sec:model} discuss the

384: model and approximations used to calculate the free energy of the

385: colour superconducting phases that are of interest here.

386: In all cases, electrical

387: and colour neutrality constraints are imposed. Section \ref{sec:results}

388: shows the behaviour of the 1SC and CSL gap parameters as functions of

389: $\mu$ and $T$. The gaps are then used in a free energy calculation,

390: the results of which are presented in Section \ref{sec:phase_diags} as

391: a sequence of ($\mu,T$) phase diagrams for neutral three flavour quark

392: matter for different values of $M_s$. Section \ref{sec:csl_summ}

393: presents conclusions and directions for further research.

394: 

395: 

396: \clearpage

397: 

398: \section{1SC and CSL pairing}

399: \label{sec:1SC/CSL}

400: 

401: The forms of zero-temperature pairing that will be compared

402: in this paper are shown in Figs.~\ref{fig:1sc_levels},

403: \ref{fig:2sc_levels}, \ref{fig:CSL_levels}, \ref{fig:2SC_CSL_levels}.

404: In the ``\onesc''

405: phase (Fig.~\ref{fig:1sc_levels}), the Fermi momenta of the three 

406: flavours are so different that only single-flavour pairing is possible.

407: Two colours of each flavour form (\thrbarA,\thrS,1,$+$)($C\ga_3$)

408: Cooper pairs (1SC$u$, 1SC$d$ and 1SC$s$).

409: The third colour of each

410: flavour forms (\sixS,\thrS,0,$+$)($C\ga_0\ga_5$) pairs. 

411: In the \twoonesc\ phase (Fig.~\ref{fig:2sc_levels}), the up and down

412: quarks are able to pair with each other in the 2SC channel (which

413: only involves two colours), but the strange quarks undergo single-flavour

414: pairing. Again, the third colour of each

415: flavour forms (\sixS,\thrS,0,$+$)($C\ga_0\ga_5$) pairs. 

416: 

417: Figures \ref{fig:CSL_levels} and \ref{fig:2SC_CSL_levels} are the

418: equivalents of Figures \ref{fig:1sc_levels} and

419: \ref{fig:2sc_levels}, where the 1SC pairing of individual

420: flavours is replaced by CSL pairing.

421: 

422: %\clearpage

423: 

424: We now give the explicit form of the pairing patterns mentioned so far.

425: The basic two-colour single-flavour spin-1 1SC pairing

426: (the ``polar phase'' in Ref.~\cite{TS1flav}), described here as

427: (\thrbarA,\thrS,1,$+$)($C\ga_3$), is

428: \begin{equation}

429:   \label{eq:1SC_ansatz}

430:   \mathbf{\De}^{ab,ij}_{\al\be} 

431:   = \De_{1SC}(C\ga_3)^{ab}(\la_2)_{\al\be}\de^{ij} \ .

432: \end{equation}

433: The colour indices are $\al,\be$, and $\la_2$ is an

434: antisymmetric Gell-Mann matrix in colour space. The Dirac indices

435: are $a,b$, and $C=i\ga_0\ga_2$ is the usual charge conjugation matrix.

436: The flavour indices $i,j$ are included for completeness: in this single-flavour

437: channel $i=j=1$.

438: This condensate breaks rotational invariance, as the explicit

439: occurrence of the $\ga_3$ matrix indicates. 

440: Only two colours (typically taken as red and green) of quark are

441: involved in the pairing meaning that the remaining (blue) quark is

442: required to find an alternative channel in which to pair. The

443: $SU(3)_{\rm colour}$ symmetry is broken down to an $SU(2)$

444: subgroup.

445: 

446: The single-colour single-flavour (\sixS,\thrS,0,$+$)($C\ga_0\ga_5$)

447: pairing is

448: \begin{equation}

449:   \mathbf{\De}^{ab,ij}_{\al\be} 

450:   = \De_{05}(C\ga_0\ga_5)^{ab}\de_{\al\be}\de^{ij} \ .

451: \end{equation}

452: This channel is attractive in an NJL model with

453: gluon-exchange-type interaction, but

454: the pairing disappears as the quark masses go to zero \cite{Alford:2002kj}.

455: This means it may be somewhat suppressed for up and down quarks,

456: depending on how much of their vacuum constituent mass survives

457: at the relevant densities \cite{BHO,Huang:2001yw,Blaschke:2003cv}.

458: Our earlier NJL study estimated the pairing gap in the $C\ga_0\ga_5$ channel 

459: to be of the order 1-100 eV. This is

460: much smaller than the typical gaps in the 2SC, 1SC, and

461: CSL channels, so when we perform free energy calculations we will

462: neglect the (\sixS,\thrS,0,$+$)($C\ga_0\ga_5$) pairing, treating

463: those quarks as unpaired.

464: 

465: CSL pairing is

466: \begin{equation}

467:   \label{eq:CSL_ansatz}

468:   \mathbf{\De}^{ab,ij}_{\al\be} 

469:    = \De_{\rm CSL}\,(C\ga_A)^{ab}\,\la^A_{\al\be}\,\de^{ij},

470: \end{equation}

471: where we sum over $A=1,2,3$, $\ga_A = (\ga_1,\ga_2,\ga_3)$ and $\la_A =

472: (\la_7,\la_5,\la_2)$ which are the antisymmetric 

473: Gell-Mann matrices in colour space. 

474: %% In colour space ($r,g,b$), the gap matrix takes the form,

475: %% \begin{equation}

476: %%   \label{eq:CSL_matrix}

477: %%   \mathbf{\De} = \De_{\rm CSL}\left( 

478: %%     \begin{array}{ccc}

479: %%       0      & -C\ga_3 & -C\ga_2 \\

480: %%       C\ga_3 &  0      & -C\ga_1 \\

481: %%       C\ga_2 &  C\ga_1 &  0

482: %%     \end{array}

483: %%     \right)\sigma_1.

484: %% \end{equation}

485: %The CSL ansatz \eqn{eq:CSL_ansatz} is symmetric in spin, implying that

486: %a spin-1 condensate is generated via this pairing channel. 

487: The symmetry breaking associated with the CSL phase is

488: \begin{equation}

489:   \label{eq:CSL_symm_break}

490:   SU(3)_{\rm colour} \times SO(3)_J \times U(1)_B \rightarrow 

491:   SO(3)_{{\rm colour}+J} \times \mathbb{Z}_2,

492: \end{equation}

493: where $SU(3)_{\rm colour}$ is gauged, $SO(3)_J$ is the 

494: rotation group and

495: $U(1)_B$ is baryon number.

496: The unbroken rotational symmetry

497: $SO(3)_{{\rm colour}+J}$ corresponds to a locking between the colour and

498: spin degrees of freedom. This can be thought of as associating a

499: direction in configuration space with a linear combination of

500: directions in colour space, i.e. red and green quarks pair in the

501: $z$-direction, red and blue quarks in the $y$-direction and green and

502: blue quarks in the $x$-direction. Therefore, unlike the polar phase,

503: CSL pairing remains isotropic. The locking is similar to that which is

504: found in the CFL phase between colour and flavour. The CSL phase is

505: the colour superconducting equivalent of the $B$-phase of superfluid

506: helium-3 \cite{helium3}.

507: 

508: In general, one would expect that the CSL condensate could involve

509: a linear combination of $C \ga_i$ and $C \si_{0i}$ with a

510: colour structure that is correlated with the spatial direction.

511: However, Table 1 in Ref.~\cite{Alford:2002kj} shows that the colour

512: antisymmetric $C\sigma_{03}$ channel gives no contribution when using

513: the mean-field approximation within an NJL model with a full gluon

514: interaction vertex.  This is precisely the interaction that we use,

515: so only the $C \ga_i$ term is included in the CSL ansatz of

516: Equation \eqn{eq:CSL_ansatz}.

517: 

518: We can calculate the binding energy of the

519: CSL channel, following Ref.~\cite{Alford:2002kj}, and compare it with the

520: results for the polar 1SC channel, taken from the same

521: reference. The result is shown in Table \ref{tab:MF_CSL},

522: where for a given condensate of gap parameter $\De$

523: the binding energy via interaction $I$ is given by

524: $E = - S_I \De^2$, so larger values of $S$ imply stronger binding.

525: $I$ can be ``inst'' (instanton interaction), ``mag'' (magnetic

526: gluon interaction), or ``elec'' (electric gluon interaction).

527: The table shows that the CSL condensate is more tightly bound

528: than the polar condensate, leading one to expect that, other

529: things being equal, CSL pairing will be favored over polar pairing.

530: 

531: \begin{table*}

532: \newcommand{\st}{\rule[-2ex]{0em}{4ex}} % Strut to help row spacing

533: \setlength{\tabcolsep}{0.25em}

534: \newlength{\instwid}

535: \newlength{\gluonwid}

536: \newlength{\magwid}

537: \settowidth{\instwid}{instanton} %{$S_{\rm inst}$}

538: \settowidth{\gluonwid}{x+$S_{\rm mag}$x}

539: \settowidth{\magwid}{mag.~only}

540: \begin{tabular}{ccccccccccccc}

541: \hline\hline

542: \st phase & colour & flavour & $j$ & parity &

543: \multicolumn{2}{c}{Dirac} &  

544: \parbox{4em}{\bc BCS\\[-0.5ex] enhance-\\[-0.5ex] ment\ec}

545:  & \parbox{\instwid}{\bc ~ \\[-0.2ex] $S_{\rm inst}$ \ec}

546:  & \parbox{\gluonwid}{

547:       \bc $S_{\rm elec}$\\[-0.2ex]+$S_{\rm mag}$ \ec}

548:  & \parbox{\magwid}{\bc ~ \\[-0.2ex] $S_{\rm mag}$ \ec } \\[-1ex]

549: \hline

550: %

551: \st CSL & \thrbarA & \thrS & $1_S$ & $+$& $C\ga_i$ & LR & $\O(1)$ 

552:   & 0 & $+96$ & $+48$ \\

553: %

554: \st ``Polar'' (1SC) & \thrbarA & \thrS & $1_S$ & $+$& $C\ga_3$ & LR & $\O(1)$ 

555:   & 0 & $+32$ & $+16$  \\

556: \hline\hline

557: \end{tabular}

558: \caption[Binding strengths of CSL channel in NJL models

559: in the mean-field approximation.]

560: {Binding strengths of CSL (1st row) and 1SC (2nd row) channels in an NJL model

561: in the mean-field approximation. The 1SC results are taken from  

562: Table 1 in \cite{Alford:2002kj}. We see that CSL is more strongly bound.

563: \label{tab:MF_CSL}

564: } 

565: \end{table*}

566: 

567: 

568: \section{Outline of Calculation}

569: \label{sec:outline}

570: 

571: % The effects of finite volume on colour superconducting condensates has

572: % been studied by the authors of \cite{Amore:2001uf}. In the case of a

573: % two flavour colour superconducting state, they found that the

574: % infinite-volume limit is a good approximation, even for small systems.

575: % In light of this, finite volume effects will be ignored here. This

576: % Section will discuss the practical application of these constraints,

577: % showing how they can be incorporated self-consistently into model

578: % calculations

579: % \cite{Shovkovy:2003uu,Steiner:2002gx,Huang:2002zd,Shovkovy:2003ce,Buballa:df}.

580: The aim of this paper is to map out the $(\mu,T)$ phase diagram 

581: for the 2SC, unpaired, \onesc, \csl, \twoonesc, and \twocsl\ phases of

582: neutral three flavour quark matter at densities relevant for compact

583: stars ($\mu\sim 500$ MeV), for different values of $M_s$.

584: The phase diagram is obtained by calculating the free energy

585: of each phase, imposing neutrality conditions and

586: equilibration under the weak interactions, and minimizing

587: the free energy with respect to the pairing strengths $\De_\rho$

588: in the various phases. In section \ref{sec:model} we will

589: discuss the NJL model within which the free energy is calculated.

590: In this section we assume that the free energy is known.

591: 

592: To obtain neutral matter we introduce chemical potentials

593: coupled to the gauged charges. 

594: (In full QCD the time-components of the gauge fields play this

595: role \cite{Gerhold:2003js,Dietrich:2003nu}, and there are no

596: gluonic contributions to the charge density \cite{Gerhold:2004sk}.)

597: The electron chemical potential $\mu_e$ couples to negative electric

598: charge. To impose colour neutrality it is sufficient to couple

599: chemical potentials $\mu_3$ and $\mu_8$ to the Cartan sub-algebra

600: of the colour gauge group, generated by

601: $T_3=\mbox{diag}(1/2,-1/2,0)$ and

602: $T_8=\mbox{diag}(1/3,1/3,-2/3)$ \cite{Buballa:2005bv}.

603: 

604: The charge densities are related to the chemical potentials via the

605: derivatives of the free energy, so the neutrality constraints are

606: \beq

607: \ba{rcl}

608: \label{neutrality}

609:  \dsp n_8 \equiv -\frac{\partial\Omega}{\partial\mu_8} &=& 0,\\[3ex]

610:  \dsp n_3 \equiv -\frac{\partial\Omega}{\partial\mu_3} &=& 0,\\[3ex]

611:  \dsp n_Q \equiv \phm\frac{\partial\Omega}{\partial\mu_e} &=& 0.

612: \ea

613: \eeq

614: 

615: The weak interaction violates flavour conservation but conserves

616: baryon number, allowing the chemical potentials of the individual

617: quark species to be written in terms of 

618: $\mu_e$, $\mu_3$, $\mu_8$ and the quark number chemical potential $\mu$,

619: \cite{Alford:2004hz}

620: \beq

621: \ba{rcl}

622:  \mu_{ru} &=& \mu - \frac{2}{3}\mu_e + \frac{1}{2}\mu_3 + \frac{1}{3}\mu_8,\\

623:  \mu_{gu} &=& \mu - \frac{2}{3}\mu_e - \frac{1}{2}\mu_3 + \frac{1}{3}\mu_8,\\

624:  \mu_{bu} &=& \mu - \frac{2}{3}\mu_e - \frac{2}{3}\mu_8,\\

625:  \mu_{rd} &=& \mu + \frac{1}{3}\mu_e + \frac{1}{2}\mu_3 + \frac{1}{3}\mu_8,\\

626:  \mu_{gd} &=& \mu + \frac{1}{3}\mu_e - \frac{1}{2}\mu_3 + \frac{1}{3}\mu_8,\\

627:  \mu_{bd} &=& \mu + \frac{1}{3}\mu_e - \frac{2}{3}\mu_8,\\

628:  \mu_{rs} &=& \mu + \frac{1}{3}\mu_e + \frac{1}{2}\mu_3 + \frac{1}{3}\mu_8 

629:  - \frac{M_s^2}{2\mu} - \frac{M_s^4}{8\mu^3},\\

630:  \mu_{gs} &=& \mu + \frac{1}{3}\mu_e - \frac{1}{2}\mu_3 + \frac{1}{3}\mu_8 

631:  - \frac{M_s^2}{2\mu} - \frac{M_s^4}{8\mu^3},\\

632:  \mu_{bs} &=& \mu + \frac{1}{3}\mu_e - \frac{2}{3}\mu_8 

633:  - \frac{M_s^2}{2\mu} - \frac{M_s^4}{8\mu^3}.

634: \ea

635:   \label{eq:chem_pots}

636: \eeq

637: The strange quark mass is included to leading order as

638: a term in the strange quark chemical potential 

639: \cite{Bedaque:2001je,Alford:2004hz}.

640: 

641: The neutrality constraints will always be implemented locally,

642: i.e.~no mixed phases of quark matter and nuclear matter

643: will be discussed \cite{Neumann:2002jm,Reddy:2004my}.

644: 

645: For each condensate $\rho$, the gap equation is found analytically by

646: differentiating the free energy with respect to the gap parameter of

647: that condensate,

648: \begin{equation}

649:   \frac{\partial\Om_{\rm total}}{\partial\De_\rho} = 0. 

650: \end{equation}

651: Finding the free energy of a phase therefore corresponds to finding

652: a stationary point of $\Om$ in the parameter space of chemical

653: potentials and gap parameters.

654: If we work in the neutral subspace, fixing $\mu_e,\mu_3,\mu_8$

655: as functions of the $\De_\rho$, then the relevant stationary point

656: will always be a minimum with respect to the gap parameters.

657: 

658: %----

659: %In the work presented here the three coupled equations will only be

660: %solved in the case where the 2SC condensate is being considered. The

661: %gap equation for the CSL$s$ or 1SC$s$ condensate can then be solved on

662: %its own, using the 2SC neutrality solutions for $\mu_e$ and $\mu_8$.

663: %Figures \ref{fig:1sc_levels} and \ref{fig:CSL_levels} show the

664: %arrangement of quark Fermi surfaces in the 1SC and CSL phases

665: %respectively. The splitting between the different flavours will be the

666: %same as in completely unpaired quark matter, corresponding to values

667: %of the chemical potentials, $\mu_e = \mu_e^{unp}$ and $\mu_8 =

668: %\mu_8^{unp}$. This implies that is is unnecessary to simultaneously

669: %solve the gap equation and neutrality conditions for the phases that

670: %only involve single-flavour pairings since the neutrality solutions

671: %are already known.

672: %----

673: 

674: We can now anticipate some simplifications that will streamline

675: our calculation.

676: 

677: Firstly, as is clear from Figs.~\ref{fig:1sc_levels},

678: \ref{fig:2sc_levels}, \ref{fig:CSL_levels}, \ref{fig:2SC_CSL_levels},

679: we will always be concerned with 

680: colour superconducting condensates that are at least invariant under

681: the $SU(2)$ colour subgroup that rotates red and green

682:  quarks into each other. This means that $\mu_3$ will always be zero.

683: 

684: Secondly, it is never necessary to solve the full set of coupled

685: gap and neutrality equations. The single-flavour pairing is so weak

686: that its back-reaction on the chemical potentials via the neutrality

687: condition can be neglected. Thus for the \twoonesc\ and \twocsl\

688: phases one can first treat the strange quarks as free, determine

689: $\mu_e$ and $\mu_8$ by solving for 2SC pairing, and then use

690: those values in the gap equations for the strange quark pairing.

691: Similarly, for the \onesc\ and \csl\ phases one can use the values

692: of $\mu_e$ and $\mu_8$ that would ensure neutrality in totally 

693: unpaired quark matter (in the case of $\mu_8$ this value is zero).

694: 

695: Actually, the only phase where the exact value of $\mu_8$ is important 

696: is the \twocsl\ phase, where the nonzero $\mu_8$

697: puts a stress on the CSLs pairing.

698: Following the standard argument \cite{Alford:2003fq}, we expect that

699: CSL pairing can occur as long as the cost of breaking a CSL pair

700: is greater than the gain in energy from converting a blue quark

701: to a red/green one, i.e.,

702: \begin{equation}

703:   \label{eq:CSL_pair_cond}

704:   |\mu_8| < 2\De_{\rm CSL}. 

705: \end{equation}

706: The phase diagrams to

707: be presented in section \ref{sec:phase_diags} will show this

708: behaviour.

709: 

710: 

711: 

712: \section{Model and approximations}

713: \label{sec:model}

714: 

715: %The detailed derivation of the thermodynamic potential $\Om$ for a

716: %colour superconducting phase using the NJL model can be found in

717: %Appendix \ref{app:thermo_pot_derivation}. This section presents the

718: %main points and final results of this derivation. 

719: The NJL-type

720: Lagrangian that will be used in the following calculations is,

721: \begin{eqnarray}

722:   \label{eq:L_NJL2}

723:   \Lag_{\rm NJL} & = & \Lag_{\rm kin} + \Lag_{\rm int}\\

724:   & = & \bar{\psi}( i \dslash - \muslash) \psi + \Lag_{\rm int},

725: \end{eqnarray}

726: where $\muslash = \mu\ga_0$ is a diagonal matrix in colour-flavour

727: space, with entries given by \eqn{eq:chem_pots}. 

728: The up and down quarks are treated as massless, and

729: the strange quark mass is only taken into account 

730: to leading order, as an effective chemical potential for strangeness.

731: The interaction we use is abstracted

732: from single-gluon exchange by replacing the gluon propagator with

733: a four-fermion coupling constant,

734: \begin{equation}

735:   \label{eq:L_int1}

736:   \Lag_{\rm int} = -\frac{3}{8}\,G

737:   (\bar{\psi}\Gamma^a_\mu\psi)(\bar{\psi}\Gamma_a^\mu\psi),

738: \end{equation}

739: where

740: \begin{equation}

741:   \label{eq:vertex}

742:   \Gamma^a_\mu = \gamma_\mu \lambda^a,

743: \end{equation}

744: where the $\lambda^a$'s are the standard Gell-Mann matrices

745: ($a=1\ldots 8$), giving the interaction the quantum numbers of single

746: gluon exchange. 

747: %The value of $G$ was chosen so as to give $\De_{2SC}(\mu=500,

748: %M_s=0) = 31.1$ MeV for a momentum cut-off of $\Lambda=600$ MeV,

749: %as in \cite{Alford:2003fq}. This corresponds to

750: %$\De_{\rm CFL}(\mu=500, M_s=0) = 25$ MeV.

751: 

752: In order to study diquark pairing and colour superconductivity, we

753: Fierz transform the interaction into the form

754: \begin{equation}

755:   \label{eq:L_int2}

756:   \Lag_{\rm int} = G \sum_\rho \alpha'_\rho 

757:   (\bar{\psi}\Gamma^{qq}_\rho\bar{\psi}^T)

758:   (\psi^T\bar{\Gamma}^{qq}_\rho\psi),

759: \end{equation}

760: where $\rho$ labels the different diquark pairing channels,

761: $\Ga^{qq}_\rho$ specifies the

762: form of the diquark pairing in channel $\rho$ and

763: $\bar{\Gamma}^{qq}_\rho$ is defined by

764: \begin{equation}

765:   \Ga^{qq}_\rho = \ga_0(\bar{\Ga}^{qq}_\rho)^\dagger\ga_0^T.

766: \end{equation}

767: 

768: Note that we sum up the terms for all the different pairing channels,

769: even though no single Fierzing of the original interaction would

770: give the sum of all of them. This is legitimate because there

771: are no cross-terms between different channels in the binding energy.

772: 

773: With the interaction expressed as in Eq.~\ref{eq:L_int2} it is

774: straightforward to do a mean field calculation of the free energy is

775: now in a suitable form to allow the free energy to be calculated. This

776: is done by performing a bosonisation of the Fierz transformed

777: Lagrangian via a Hubbard-Stratonovich transformation

778: \cite{Kikkawa:1976fe,Ebert:1991pz,Reinhardt:1989rw,Berges:1998rc}.

779: % Following the derivation in the Appendix

780: % \ref{app:thermo_pot_derivation}, 

781: The general expression for the

782: thermodynamic potential of a colour superconducting condensate is

783: \begin{equation}

784:   \label{eq:thermo_pot_eqn}

785:   \Omega = -T \sum_n \int 

786:   \frac{d^3p}{(2\pi)^3}\,\frac{1}{2}{\rm Tr\, ln}

787:   \left(\frac{S^{-1}(i\omega_n,\vec p)}{T}\right) 

788:   + \frac{\De_\rho\De_\rho}{4\alpha^\prime G},

789: \end{equation}

790: where $S^{-1}(i\omega_n,p)$

791: %% \begin{equation}

792: %%   S^{-1}(\pbar) = \left(\begin{tabular}{cc}

793: %%       $\bar{\pslash} -  \mathbf{m} + \muslash$ & $\Gamma_{qq}\De$ \\

794: %%       $\De\bar{\Gamma}_{qq}$ & $(\bar{\pslash} +  \mathbf{m} - \muslash)^T$

795: %%     \end{tabular}\right),

796: %% \end{equation}

797: is the full inverse quark propagator.

798: %(see Equations \eqn{eq:NG_basis} and \eqn{eq:gap_ansatz}). 

799: $\omega_n=(2n-1)\pi T$ are the Matsubara frequencies and the trace

800: is over colour, flavour, and spinor indices.

801: The Matsubara summation is performed using the identity

802: \begin{equation}

803:   \label{eq:matsubara1}

804:   T\sum_n\ln\left(\frac{\omega_n^2 + \eps_a(\vec p)^2)}{T^2}\right)

805: = |\eps_a(\vec p)| + 2T\ln(1 + e^{-|\eps_a(\vec p)|/T}).

806: \end{equation}

807: %% Upon performing the trace, the free energy density can be written as,

808: %% \begin{equation}

809: %%   \label{eq:thermo_pot}

810: %%   \Om = \Om_0 + \frac{\De^2}{4\al^\prime G}

811: %%   - \bar{N_f}\bar{N_c}\sum_a \int\frac{d^3p}{(2\pi)^3}

812: %%   \left(\eps_a(p) + 2T\ln\left(1+e^{-\eps_a(p)/T}\right)\right).

813: %% \end{equation}

814: %% $\al^\prime$ is a constant and comes from performing a

815: %% Fierz transformation on the Lagrangian, \eqn{eq:L_int1}. The origin

816: %% of $\al^\prime$ will be discussed in greater detail in The

817: %% degeneracy factors $\bar{N_f}$ and $\bar{N_c}$ appear when taking the

818: %% trace of the quark propagator and correspond to the number of flavours

819: %% and colours of quarks involved in a particular pairing. They arise as

820: %% a result of the particular gap ansatz that is being studied and will

821: %% change depending upon the phase of interest. 

822: The functions $\eps_a(p)$ are the dispersion relations for the

823: fermionic quasiparticles.

824: They are not explicitly $T$-dependent, but

825: they depend upon the gap and quark chemical potentials which are

826: $T$-dependent. 

827: 

828: For convenience, we use neutral unpaired quark

829: matter as the zero of free energy, so we subtract it from the

830: value computed above. This gives a physically meaningful quantity

831: without ultraviolet divergences:

832: \beq

833: \ba{r@{~}c@{~}l}

834:   \label{eq:FE_Tneq0}

835:   \Om =  &-&\dsp \int \frac{d^3p}{(2\pi)^3}\,\frac{1}{2}

836:   \sum_a \left(|\eps_a(\vec p)| 

837:      + 2T\ln(1 + e^{-|\eps_a(\vec p)|/T})\right)\\[2ex]

838:   &+&\dsp \frac{\De_\rho\De_\rho}{4\alpha^\prime G}

839:    - \frac{1}{12\pi^2}\left(\mu_e^4 + 2\pi^2T^2\mu_e^2 

840:      + \frac{7\pi^4}{15}T^4\right) \\[2ex]

841:   &-&\Om_{\rm free}(\mu,M_s,\mu^{\rm unp}_e,T) 

842: \ea

843: \eeq

844: where the electron contribution to the free energy has been included.

845: The electron chemical potential in

846: neutral unpaired quark matter is

847: \begin{equation}

848:   \label{eq:neutral_sols}

849:   \mu_e^{unp} = \frac{M_s^2}{4\mu} \ .

850: \end{equation}

851: The quasiquark dispersion relations $\eps_a(\vec p)$

852: are the values of the energy at

853: which the propagator diverges, i.e.

854: \begin{equation}

855:   \label{eq:detS}

856:   \mathrm{det}\,S^{-1}(\eps_a(\vec p),\vec p) = 0. 

857: \end{equation}

858: The calculation of $\Om$ then depends upon the calculation of the

859: determinant of the matrix $S^{-1}$. This matrix can be block-diagonalized

860: in the colour-flavour space, with one block for each independent pairing

861: channel. For each colour and flavour there are eight dispersion relations

862: (four Dirac branches including left-handed and right-handed, 

863: particle and antiparticle, doubled by the Nambu-Gorkov formalism).

864: In calculating the free energy of quark matter with three flavours

865: and three colours the index $a$ in \eqn{eq:detS} will therefore

866: vary from 1 to 72.

867: 

868: \subsection{Dispersion relations}

869: \label{sec:disp_rels}

870: 

871: For each unpaired colour and flavour of quark, there two branches

872: of the dispersion relation (particle and antiparticle) each with

873: multiplicity 4:

874: \beq

875: E^\pm_{\rm free} = p\pm\mu

876: \eeq

877: 

878: For each condensate we now give the dispersion relation

879: of the quasiquarks (with their multiplicity)

880: and the binding energy parameter $\al'$.

881: 

882: \subsubsection{2SC}

883: For 2SC pairing of the red and green quarks

884: of up and down flavour, the dispersion relations are

885: \beq

886:   E_{\rm 2SC} =

887:   \sqrt{(p \pm \bar{\mu})^2 + \De^2} \pm \de\mu

888: \eeq

889: each with multiplicity 4 (spin and Nambu-Gorkov), where

890: \beq

891: \ba{rcl}

892:   \bar{\mu} &=& \dsp \frac{1}{2}(\mu_{ru} + \mu_{gd}) = \frac{1}{2}(\mu_{ru} + \mu_{gd})\\ 

893:   &=& \dsp \mu - \frac{\mu_e}{6} + \frac{\mu_8}{3},\\

894:   \de\mu &=& \dsp \frac{1}{2}(\mu_{gd} - \mu_{ru}) = \frac{1}{2}(\mu_{rd} - \mu_{gu}) 

895:   = \frac{\mu_e}{2}\ .

896: \ea

897: \eeq

898: The binding energy parameter is

899: \begin{equation}

900:   \al^\prime = \frac{1}{4}.

901: \end{equation}

902: If $\De_{2SC} < \de\mu$, then the 2SC condensate is in the gapless 2SC

903: state \cite{Shovkovy:2003uu,Huang:2003xd}.

904: 

905: \subsubsection{1SC}

906: For 1SC pairing of red and green quarks of some flavour $f$

907: (assumed to have common chemical potential $\bar\mu$),

908: the dispersion relations are

909: \begin{equation}

910:   E_{\rm 1SC}^2 = 

911:   p^2 + m^2 + \bar{\mu}^2 + \De^2

912:  \pm 2\sqrt{ \bar{\mu}^2(p^2+m^2) +  \De^2 \,p_3^2} \ ,

913: \end{equation}

914: each with multiplicity 4. The binding energy parameter is

915: \begin{equation}

916:   \al^\prime = \frac{1}{8}.

917: \end{equation}

918: 

919: \subsubsection{CSL}

920: For CSL pairing of three colours of a single massless flavour,

921: in the case $\mu_8=0$, we can obtain a simple closed-form expression for

922: the determinant of the inverse propagator,

923: \beq

924: \label{eq:CSLdet}

925: \ba{rcl}

926:  \multicolumn{3}{l}{\det\,S^{-1}(p_0,\,p) =} \\

927:   &&(\De^2 + \mu^2 - (p - p_0)^2)^2 \\

928:   &\times& (\De^2 + \mu^2 - (p + p_0)^2)^2 \\

929:   &\times& \Bigl(4 \De^4 + \mu^4 + (p^2 - p_0^2)^2 - 2 \mu^2 (p^2 + p_0^2)\\

930:   &&  + \De^2 (5 \mu^2 + (3 p - 5 p_0) (p + p_0))\Bigr)^2 \\

931:   &\times& \Bigl(4 \De^4 + \mu^4 + (p^2 - p_0^2)^2 - 2 \mu^2 (p^2 + p_0^2) \\ 

932:   &&  + \De^2 (5 \mu^2 + (3 p + 5 p_0) (p - p_0))\Bigr)^2 \ ,

933: \ea

934: \eeq

935: which has dimension 24, as expected.

936: From this the dispersion relations of the 24 quasiparticles can be obtained.

937: Note that there are 4 gapless modes (first line of the determinant),

938: and the remaining 20 are gapped. When we include a nonzero quark mass

939: all the quark modes are gapped, which is different from the

940: behaviour in weak-coupling QCD \cite{Schmitt:2004et}.

941: For a discussion of this point see Refs. \cite{Aguilera:2005tg,Schmitt:2005wg}.

942: In the general case $\mu_8\neq 0, m>0$,

943: the determinant can be readily obtained using a symbolic mathematics

944: program such as {\em Mathematica}, but is too complicated to display here.

945: 

946: The binding energy parameter is

947: \beq

948:   \al^\prime = \frac{1}{8}.  

949: \eeq

950: 

951: 

952: \section{Results}

953: \label{sec:results}

954: 

955: The following sections present the results of solving the gap and neutrality

956: equations and calculating the free energies for the 

957: \onesc, \csl, \twoonesc, and \twocsl\ condensates.

958: The momentum cutoff was fixed at $\Lambda=600$~MeV, with

959: the value of the coupling $G$ calibrated to give $\De_{2SC}(\mu=500,

960: M_s=0) = 31.1$~MeV, as in Ref.~\cite{Alford:2003fq}. This corresponds to

961: $\De_{\rm CFL}(\mu=500, M_s=0) = 25$ MeV. We also performed some calculations

962: at $\La=800~\MeV$ with the same calibration, to check for cutoff sensitivity.

963: 

964: If single-flavour pairing were ignored, the phase diagram would

965: contain a transition from unpaired quark matter 

966: to 2SC quark matter at $\mu = M_s^2/(2\mu)$. In principle this

967: critical value also depends on temperature, but at the

968: temperatures of interest to us ($T\lesssim 1~\MeV$) this

969: can be neglected.

970: Thinking of this as a background, we can then include the

971: possibility of 1SC and CSL single-flavour pairing, and we expect

972: to find that in the low-temperature parts of the unpaired region

973: there are regions of \onesc\ or \csl\ phases, and in the

974: low-temperature parts of the 2SC region there are regions of \twocsl\

975: or \twoonesc. This is indeed what we ultimately find.

976: 

977: As mentioned in section \ref{sec:model}, we can calculate $\mu_e$

978: and $\mu_8$ in the 2SC/unpaired background, before including the

979: effects of single flavour pairing. We show these results 

980: in the next subsection. Then we go on to solve the gap equations for

981: the various forms of single flavour pairing, and compare their

982: free energies.

983: 

984: \subsection{Chemical potentials with no single flavour pairing}

985: \label{sec:chem_pots}

986: 

987: Figure \ref{fig:mue_mu8_mu} shows (for four different strange quark

988: masses) the values of $\mu_e$ and $\mu_8$ in neutral

989: unpaired/2SC matter at $T\approx 0$

990: over a range of quark chemical potential $450~\MeV < \mu < 550~\MeV$.

991: This required simultaneously solving the 2SC gap equation

992: and the two neutrality equations, and choosing the phase with

993: the lowest free energy at each value of $\mu$.

994: 

995: At $\mu\approx M_s^2/(2\mu)$

996: we see the transition from unpaired to 2SC, at which

997: $\mu_e$ jumps from about $M_s^2/(4 \mu)$ to about $M_s^2/(2\mu)$

998: \cite{AR-02}. In the unpaired phase,

999: $\mu_8$ is zero because nothing picks out any direction in colour space.

1000: In the 2SC phase, however, the red and green light quarks pair, and

1001: this means that a nonzero $\mu_8$ is required to maintain colour

1002: neutrality. At moderate densities $\mu_8$ is negative

1003: because the number of red and green quarks is enhanced

1004: by the 2SC pairing (there is more phase space above the Fermi

1005: surface than below it), and a negative value of $\mu_8$ restores

1006: the colour balance by favoring blue quarks ($T_8={\rm diag}(1,1,-2)$).

1007: As the quark chemical potential approaches the cutoff, however,

1008: the situation is reversed: degrees of freedom above the cutoff

1009: are excluded, so the condensate contains more red and green holes

1010: below the Fermi surface than red and green quarks above it.

1011: Thus $\mu_8$ turns around and starts to rise again as $\mu$ approaches

1012: $\La$. Depending on the values of the strange quark mass

1013: and the cutoff, $\mu_8$ may cross through zero and become positive

1014: at some specific density, as seen in Fig.~\ref{fig:mue_mu8_mu}.

1015: This is a very interesting phenomenon, because it leads to an

1016: ``island'' of \twocsl\ pairing in the phase diagram

1017: (Fig.~\ref{fig:phase_diags}),

1018: but we emphasize that it is sensitive to the value of the cutoff, and

1019: if we work at $\La=800~\MeV$ then we find that $\mu_8$ 

1020: only becomes positive at $\mu>\La$.

1021: 

1022: \begin{figure}[t]

1023:   \includegraphics[scale=0.36]{./figures/mue_mu_T0.03.eps}\\[1cm]

1024:   \includegraphics[scale=0.36]{./figures/mu8_mu_T0.03.eps}

1025:   \caption{\label{fig:mue_mu8_mu} 

1026: For 2SC/unpaired quark matter, the variation of  $\mu_8$ 

1027: with quark chemical potential at $T\approx 0$ and for various values 

1028: of $M_s$. We see the transition from unpaired ($\mu_8=0$) at

1029: low $\mu$ to 2SC at high $\mu$.

1030:   }

1031: \end{figure}

1032: 

1033: \subsection{Gap parameter plots}

1034: \label{sec:gap_plots}

1035: 

1036: We now proceed to solve the gap equations for the single-flavour-pairing

1037: phases.

1038: Figure \ref{fig:1SC_CSL_gaps_mu} shows how the single-flavour pairing

1039: gap parameters 

1040: vary over the quark chemical potential range  $450~\MeV < \mu < 550~\MeV$.

1041: In all cases, the gaps for the different flavours follow the same

1042: pattern. 

1043: 

1044: For the purely single-flavour phases (\onesc\ and \csl)

1045: there are three gap parameters, for up, down, and strange quarks

1046: (red and green only, see Figs.~\ref{fig:1sc_levels},\ref{fig:CSL_levels}).

1047: These are shown by

1048: the various broken lines in Fig.~\ref{fig:1SC_CSL_gaps_mu}.

1049: They exhibit the hierarchy,

1050: \begin{equation}

1051:   \label{eq:single_flav_order_mu}

1052:   \De^d > \De^u > \De^s.

1053: \end{equation}

1054: This follows from the electrical-neutrality-induced ordering

1055: of Fermi momenta in unpaired quark matter

1056: \mbox{$p_F^d > p_F^u > p_F^s$}

1057: and the fact that quarks with larger Fermi momentum have more phase space 

1058: at their Fermi surface, and hence stronger pairing.

1059: We also see that 1SC pairing gives a larger gap than CSL pairing,

1060: but it will turn out that CSL has a larger condensate binding energy

1061: (see Table \ref{tab:MF_CSL}) and so, other thing being equal, 

1062: we expect CSL to be favoured over

1063: 1SC. This is confirmed by the free energy calculation and can be

1064: seen by the presence of the CSL phase in the phase diagrams of section

1065: \ref{sec:phase_diags}.

1066: 

1067: In the \twocsl\ and \twoonesc\ phases only the strange quarks

1068: undergo single-flavour pairing, so there is only one

1069: gap parameter, $\De^s$. This is shown

1070: by the solid lines in Fig.~\ref{fig:1SC_CSL_gaps_mu}).

1071: In the top left panel of the figure, the solid line 

1072: (\twoonesc) ends at $\mu\approx 510~\MeV$, because the 2SC phase

1073: ceases to be favored below that point. 

1074: Notice that in general the value of $\De^s$ in the presence of

1075: 2SC pairing of the up and down quarks is slightly higher than its

1076: value when all three flavours undergo single-flavour pairing.

1077: This is because 2SC pairing pulls the up and down Fermi surfaces

1078: together, which drags the strange Fermi momentum up a little

1079: in order to maintain electrical neutrality. This increases the phase space

1080: at the strange quark Fermi surface, boosting the gap by a small amount.

1081: 

1082: The gap parameters for the CSL$s$ condensate in the \twocsl\ phase 

1083: show a non-monotonic behaviour (solid lines in lower two panels of

1084: Fig.~\ref{fig:1SC_CSL_gaps_mu}). That is because CSL pairing

1085: is very sensitive to the value of the $\mu_8$ chemical potential,

1086: which tries to split apart the Fermi momentum of the

1087: blue quarks from those of the red and green quarks, and the

1088: background 2SC pairing induces a non-zero $\mu_8$ to

1089: maintain colour neutrality. CSL pairing

1090: is disfavored when its gap is less than a critical value

1091: of order $\mu_8$. For the cutoff that we used, $\La=600~\MeV$,

1092: $\mu_8$ goes through zero at $\mu\sim 530~\MeV$

1093: (see Sect.~\ref{sec:chem_pots}), so the CSL gap is boosted there.

1094: For a different cutoff, this would have occurred at a different density.

1095: 

1096: % Title: 148

1097: % subtitle: 130

1098: \begin{figure}[t]

1099:   \includegraphics[scale=0.36]{./figures/1SC_gaps_T0.03.eps}

1100:   \\[0.5cm]

1101:   \includegraphics[scale=0.36]{./figures/CSL_gaps_T0.03.eps}

1102:   \caption{\label{fig:1SC_CSL_gaps_mu}

1103:     Variation of the 1SC (upper) and CSL (lower) gaps with quark

1104:     chemical potential at $T=0.03$ MeV and for $M_s=250,175$ MeV.

1105:     Solid lines are for the strange quarks, in the case where up and down

1106: quarks are undergoing 2SC pairing. The various broken lines are for

1107:     the three flavours of quark in the case where all of them undergo

1108: single-flavour pairing.}

1109: \end{figure}

1110: 

1111: \subsection{Non-BCS temperature dependence}

1112: \label{sec:non_BCS_temp}

1113: 

1114: \begin{figure}[t]

1115:   \includegraphics[scale=0.36]{./figures/1SC_gaps_mu540.eps}\\[0.5cm]

1116:   \includegraphics[scale=0.36]{./figures/CSL_gaps_mu540.eps}

1117:   \caption{Variation of the 1SC (upper)

1118:     and CSL (lower) gaps with temperature at $\mu=540$ MeV and for

1119:     $M_s=250,175$ MeV. 

1120:     Solid lines are for the strange quarks, in the case where up and down

1121: quarks are undergoing 2SC pairing. The various broken lines are for

1122:     the three flavours of quark in the case where all of them 

1123: potentially undergo single-flavour pairing.}

1124:   \label{fig:1SC_CSL_gaps_T}

1125: \end{figure}

1126: 

1127: The temperature dependence of the single-flavour gap

1128: parameters is shown in

1129: Fig.~\ref{fig:1SC_CSL_gaps_T}. We see that the gap

1130: parameters drop with

1131: increasing temperature, and that each vanishes at a second-order

1132: phase transition at some critical temperature $T_{\rm crit}$.

1133: The critical temperatures deviate slightly from the BCS prediction

1134: $T_{\rm crit} \simeq 0.57\, \De(T=0)$: 

1135: \begin{eqnarray}

1136:   \label{eq:CSL_T_dep}

1137: \mbox{1SC}:\quad  T^f_{crit} &\approx& 0.51\, \De^f_{1SC}(T=0),\\

1138: \mbox{CSL}:\quad  T^f_{crit} &\approx& 0.82\, \De^f_{\rm CSL}(T=0).

1139: \end{eqnarray}

1140: Such deviation was also found

1141: in weak-coupling-QCD studies of 

1142: single-flavour pairing \cite{Schmitt:2002sc}, and

1143: non-BCS critical temperatures have also been seen

1144: in gapless phases \cite{Huang:2003xd}

1145: and crystalline phases \cite{Bowers:2001ip}.

1146: 

1147: % title: 150

1148: % axis labels: 130

1149: % tick labels: 123

1150: % phase labels: 130

1151: \begin{figure*}[t]

1152:   \includegraphics[scale=0.36]{./figures/phase_diagram_Ms250.eps}

1153:   \includegraphics[scale=0.36]{./figures/phase_diagram_Ms225.eps}\\[2ex]

1154:   \includegraphics[scale=0.36]{./figures/phase_diagram_Ms200.eps}

1155:   \includegraphics[scale=0.36]{./figures/phase_diagram_Ms175.eps}

1156:   \caption{Phase diagrams for one- and two- flavour pairing in

1157:   neutral three flavour quark matter with values of $M_s$ between

1158:  175 and 250 MeV, calculated in an NJ model with cutoff $\La=600~\MeV$.

1159:   For the \twoonesc\ and \twocsl\ phases see Figs.~\ref{fig:2sc_levels}

1160:   and \ref{fig:2SC_CSL_levels}. The phase labelled ``CSL'' involves

1161:   CSL pairing of some or all of the $u$,$d$, and $s$ quarks.

1162:   }

1163:   \label{fig:phase_diags}

1164: \end{figure*}

1165: 

1166: 

1167: It is interesting to note that because each flavour has a different

1168: critical temperature for its self-pairing, we expect a series of

1169: phase transitions as the system cools (imagining these phases 

1170: to be in the core of a neutron star, for example), with single-flavour

1171: pairing appearing first for the down quarks, then for the up quarks,

1172: and finally for the strange quarks.

1173: 

1174: In the absence of any 2SC pairing, we also see that the critical temperature

1175: for 1SC pairing of any given flavour is approximately equal to

1176: the critical temperature for CSL pairing of the same flavour, 

1177: even though the 1SC phase has a larger gap. This means that

1178: if we heat up a \csl\ phase, we do not expect it to turn into

1179: \onesc.

1180: However, in the presence of 2SC pairing, we find that the critical

1181: temperature for \twocsl\ is smaller than that for \twoonesc,

1182: so we expect to find \twocsl$\to$\twoonesc$\to$2SC as we

1183: heat up the system, and this is seen in the phase diagrams

1184: of section \ref{sec:phase_diags}.

1185: 

1186: We repeated our calculations for a higher cutoff $\Lambda=800$~MeV,

1187: calibrating the coupling to give the same 2SC gap, and we found that

1188: the single-flavour pairing gaps dropped by a factor of $\sim 8$,

1189: but the relations \eqn{eq:CSL_T_dep} were still valid.

1190: 

1191: 

1192: \section{Discussion of phase diagrams}

1193: \label{sec:phase_diags}

1194: 

1195: Section \ref{sec:model} describes how we calculate the

1196: free energy for each phase as a function of ($\mu, T, M_s$). 

1197: At each value of ($\mu$,$T$) the

1198: favored phase is the one with the lowest free energy.

1199: Fig. \ref{fig:phase_diags} presents this information as

1200: a series of phase diagrams.

1201: 

1202: The crude structure of the diagrams involves

1203: the expected transition from unpaired quark

1204: matter at low density to two-flavour-paired quark matter (2SC) at

1205: high density, at a critical chemical potential $\mu_{\rm crit} 

1206: \approx M_s^2/(2\mu)$. In the $\mu$-$T$ diagram this transition 

1207: occurs along a line that comes down very steeply at $\mu=\mu_{\rm crit}$,

1208: because we are interested in low temperatures $T\lesssim 1~\MeV$,

1209: so $T\ll\De_{2SC}$. As the strange quark mass drops, this line

1210: moves to the left, and 2SC occupies more and more of the phase space.

1211: 

1212: In both the unpaired and 2SC regions of phase space, we find

1213: that, at sufficiently low temperatures, single-flavour

1214: pairing occurs. Below $\mu_{\rm crit}$

1215: at zero temperature, the quark matter is in the \csl\ phase

1216: in which all three flavours undergo CSL pairing, as depicted in

1217: Fig.~\ref{fig:CSL_levels}.

1218: % which is the favored pairing pattern

1219: % for a single flavour with no colour chemical potentials.

1220: Since the three flavours have different Fermi momenta,

1221: they have different CSL gaps, and different critical temperatures,

1222: so the region we have labelled in the phase diagrams

1223: as ``CSL'' actually contains three

1224: bands, one where just down quarks undergo CSL pairing, then below that

1225: one where down and up quarks undergo CSL pairing, and finally

1226: at the lowest temperatures one where all three flavours

1227: undergo CSL pairing. We do not expect any 1SC pairing in this region

1228: because the 1SC phases do not have higher critical temperatures

1229: than the CSL phases (Fig.~\ref{fig:1SC_CSL_gaps_T}).

1230: 

1231: Above $\mu_{\rm crit}$, the phase structure is more complicated.

1232: A colour chemical potential $\mu_8$ is generated in order to keep

1233: the 2SC phase colour neutral, and this acts to 

1234: split the blue quark Fermi momentum apart from the red and green quark

1235: Fermi momenta. The value of $\mu_8$ is typically of order a few

1236: MeV, which is easily large enough to disrupt any attempt at

1237: CSL pairing, which itself only has a gap on the order of one MeV.

1238: As a result, the favored phase at low temperature for

1239: $\mu>\mu_{\rm crit}$ is typically the \twoonesc\ phase, in which the red and

1240: green strange quarks form a spin-1 condensate (Fig.~\ref{fig:2sc_levels}).

1241: At a particular value of $\mu$, however, $\mu_8$ may pass through

1242: zero, and in that region of phase space, where $|\mu_8|< 2\De_{\rm CSL}$,

1243: the strange quarks of all three colours are able to pair in a \twocsl\ phase

1244: (Fig.~\ref{fig:2SC_CSL_levels}), which produces an ``island'' of

1245: \twocsl\ in the \twoonesc\ band.

1246: For our particular choice of NJL

1247: model and cutoff, this occurs at $\mu\approx 530~\MeV$

1248: for $M_s=175,\, 200,\, 225~\MeV$. As we emphasized in section 

1249: Sect.~\ref{sec:chem_pots}, however, the position of such an

1250: island is sensitive to details of the cutoff, and is not a

1251: robust prediction of the model.

1252: 

1253: \section{Conclusions}

1254: \label{sec:csl_summ}

1255: 

1256: In this paper we have shown that single flavour pairing

1257: is strongly affected by what is happening to other flavours.

1258: In particular, two-flavour pairing of the light flavours

1259: induces large enough colour chemical potentials to change

1260: the nature of the favored single-flavour pairing pattern

1261: for the strange quarks.

1262: Our results, shown in Fig.~\ref{fig:phase_diags},

1263: were obtained using a NJL model

1264: in the mean field approximation

1265: with a pointlike four-fermion interaction whose

1266: index structure is based on single gluon exchange.

1267: We treated the up and down quarks as massless, and treated the

1268: strange quark mass as a fixed parameter.

1269: Some of the interesting features, such as the \twocsl\ ``island'',

1270: are cutoff-dependent, but we found that

1271: the basic structure of the diagram is a robust prediction

1272: of the model. At high temperatures ($T\gg 1~\MeV$), only flavour-antisymmetric

1273: pairing such as 2SC can survive. At

1274: sufficiently low temperatures we find either CSL pairing of

1275: some or all flavours,

1276: or \twoonesc\ pairing. The critical temperature for the formation of these

1277: condensates is not reliably predicted by our model, but appears to

1278: be of order 1 MeV.

1279: 

1280: Our analysis was focused on the 2SC/1SC/CSL comparison.

1281: We ignored other single-flavour pairing patterns

1282: such as those with Dirac structure $C\si_{03}$ and $C\ga_0\ga_5$, since for 

1283: an NJL model with our interaction

1284: they have much smaller gap parameters, and hence either make a negligible

1285: contribution to the free energy, or are

1286: disfavored relative to the 1SC ($C\ga_3$) phase \cite{Alford:2002kj}.

1287: (It is interesting to note that weak-coupling QCD calculations

1288: find that a transverse {\em planar} phase is the strongest competitor

1289: to CSL \cite{Schmitt:2004et}; this phase has not

1290: yet been studied in the NJL context.)

1291: More importantly, we also 

1292: explicitly ignored other possible competing phases such as the

1293: CFL phase and crystalline (LOFF) pairing, and we also

1294: ignored chiral condensates

1295: that would give larger (density dependent) quark masses.

1296: One natural direction in which to extend this work is to perform

1297: a simultaneous comparison of all known phases, along the lines of

1298: Refs.~\cite{Ruster:2005jc,Blaschke:2005uj}, although a definitive calculation

1299: is not yet possible because we still do not know what true ground

1300: state underlies the unstable gapless phases 

1301: \cite{Huang:2003xd,Alford:2003fq,Huang:2004am,Casalbuoni:2004tb,Giannakis:2004pf,Fukushima:2005cm}.

1302: 

1303: 

1304: Another topic for future work is the possible ramifications of

1305: our results for the phenomenology of quark matter in neutron stars.

1306: The phase diagrams of Fig.~\ref{fig:phase_diags} show that a core

1307: of quark matter inside a neutron star could have a complicated

1308: radial structure of different phases, each with its own

1309: special properties. Moreover, this structure would change over time

1310: as the star's interior cooled. In our calculation the single-flavour

1311: phases had gaps of order 1~MeV, so the phase structure would settle down

1312: less than a year after the supernova,

1313: but it should be noted that we completely ignored

1314: the very weak single-flavour single-colour pairing that accompanies

1315: most of these phases, whose gap parameter could be as small as 1 eV

1316: \cite{Alford:2002kj}, leading to the emergence of further phase structure

1317: after millions of years.

1318: 

1319: For neutron star phenomenology, we are interested in the transport properties 

1320: of the different phases, in particular their permeability to magnetic

1321: fields, as well as thermal properties, neutrino emissivity and 

1322: mean free path, viscosity, and so on. These are sensitive to the

1323: low-energy degrees of freedom, such as Goldstone bosons and

1324: ungapped quark modes. 

1325: We postpone a detailed study

1326: and now sketch some of the obvious features.

1327: %The CSL-type and 1SC-type phases have quite different properties

1328: %from those of unpaired or 2SC quark matter.

1329: First, some general points:

1330: a massive flavour with CSL pairing is completely 

1331: and isotropically gapped with an energy gap of approximately

1332: $m\Delta/\mu$ for $m,\De\ll\mu$, so some of the quasiquarks become

1333: gapless in the massless limit (see discussion after Eq.~\eqn{eq:CSLdet}).

1334: A massive flavor with 1SC pairing has a direction-dependent gap, and in the

1335: massless limit becomes ungapped at two points on the Fermi surface.

1336: However we must remember that, like 2SC pairing,

1337: 1SC pairing only involves two of the three colours, so there may

1338: be other gapless modes involving the third colour.

1339: %

1340: Focussing on the phases that we have discussed here,

1341: the most completely gapped phase is \csl, in which all quark modes

1342: are gapped, although if the up and down flavours have low

1343: constituent masses there will be modes with rather small gaps.

1344: (From Ref.~\cite{Ruster:2005jc}, Fig.~4, we expect $m_u,m_d\sim 20~\MeV$

1345: at $\mu\approx 400~\MeV$, so assuming a CSL gap parameter of

1346: 1~\MeV\ we find an energy gap of about $50~\keV$.)

1347: Next comes the \twocsl\ phase, which 

1348: has blue up and down quarks that are effectively

1349: gapless (they may have an eV-scale gap due to pairing in some 

1350: very weak channel such as

1351: $C\ga_0\ga_5$, which is suppressed for light quarks). 

1352: Finally, the \twoonesc\ phase has in addition the blue strange

1353: quarks (which can generate a bigger $C\ga_0\ga_5$ self-pairing gap

1354: because the strange quark is heavier). 

1355: The \twoonesc\ phase is also special because

1356: rotational invariance is

1357: broken by the 1SC pairing, and for the red and green strange quarks

1358: the gap varies over the Fermi surface, but is always of order 

1359: $\De_{\rm 1SC}^s$ \cite{Alford:2002kj}.

1360: The full cooling phenomenology

1361: of these phases has not yet been worked out, although

1362: such questions have already been discussed

1363: for isolated single flavours \cite{Schmitt:2005wg} and for

1364: two-flavour quark matter \cite{Aguilera:2005tg}.

1365: 

1366: The magnetic properties of these phases are also expected to show

1367: some variety.

1368: In unpaired quark matter there is no Meissner effect,

1369: so magnetic flux is not expelled or confined to flux tubes.

1370: In an idealized 2SC phase the four participating quark species

1371: are uniformly gapped, but the remainder are gapless.

1372: The colour $SU(3)$ and electromagnetic $U(1)$ gauge symmetries are

1373: broken down to a colour $SU(2)$ (giving masses to five of the gluons)

1374: and a rotated electromagnetic $U(1)_{\tilde Q}$.

1375: In the \csl\ phase all the gauge bosons have Meissner masses. 

1376: The gluon masses are of order $g \De_{\rm CSL}$, and the photon mass

1377: is of order $e \De_{\rm CSL}$ \cite{Schmitt:2003xq,Schmitt:2003aa}. The

1378: magnetic properties of the \twocsl\ phase have not yet been

1379: calculated: we expect mixing of the photon and the eighth gluon,

1380: with Meissner masses that depend on $g$ as well as $e$.

1381: In the \twoonesc\ phase, under the $U(1)_{\tilde Q}$

1382: which survived the formation of the 2SC condensate,

1383: the red, green, and blue strange quarks have charges

1384: $-1,0,0$ respectively, so the additional 1SCs pairing

1385: breaks the $U(1)_{\tilde Q}$ symmetry, giving the $\tilde Q$ photon

1386: a mass of order $\tilde e\De_{\rm 1SC}^s$. The gluons have 

1387: much larger masses of order $g \De_{\rm 2SC}$. It remains to be seen whether

1388: this phase is a type-I or type-II superconductor for the

1389: $\tilde Q$ photon.

1390: 

1391: We conclude that although single-flavour pairing makes a very small

1392: contribution to the free energy of quark matter, it contributes

1393: interesting extra structure to the low-temperature part of

1394: the quark matter phase diagram. Single-flavour pairing determines

1395: the symmetries and low-energy-excitations in this region, making

1396: it directly relevant for the study of signatures of quark matter

1397: in neutron stars.

1398: 

1399: 

1400: \bc{\bf Acknowledgements}\ec

1401: This research was supported by the U.K. Particle Physics

1402: and Astronomy Research Council and by the U.S. Department of Energy

1403: under contracts DE-FG02-91ER50628 and DE-FG01-04ER0225 (OJI).

1404: We thank Andreas Schmitt for helpful comments and discussions.

1405: 

1406: 

1407: \begin{thebibliography}{99}

1408: 

1409: \bibitem{BarroisPhD}

1410: %\bibitem{Barrois:1979pv}

1411:   B.~C.~Barrois,

1412:   ``Nonperturbative Effects In Dense Quark Matter,'' Ph.D. thesis

1413:   UMI 79-04847.

1414: %\href{http://www.slac.stanford.edu/spires/find/hep/www?r=umi\%2F79-04847}{SPIRES entry} 

1415: 

1416: \bibitem{Bailin:1983bm}

1417:   D.~Bailin and A.~Love,

1418:   %``Superfluidity And Superconductivity In Relativistic Fermion Systems,''

1419:   Phys.\ Rept.\  {\bf 107}, 325 (1984).

1420:   %%CITATION = PRPLC,107,325;%%

1421: 

1422: \bibitem{ARW2}

1423:   M.~G.~Alford, K.~Rajagopal and F.~Wilczek,

1424:   %``QCD at finite baryon density: Nucleon droplets and color

1425:   %superconductivity,''

1426:   Phys.\ Lett.\ B {\bf 422}, 247 (1998)

1427:   [arXiv:hep-ph/9711395].

1428:   %%CITATION = HEP-PH 9711395;%%

1429: 

1430: \bibitem{Rapp:1997zu}

1431:   R.~Rapp, T.~Schafer, E.~V.~Shuryak and M.~Velkovsky,

1432:   %``Diquark Bose condensates in high density matter and instantons,''

1433:   Phys.\ Rev.\ Lett.\  {\bf 81}, 53 (1998)

1434:   [arXiv:hep-ph/9711396].

1435:   %%CITATION = HEP-PH 9711396;%%

1436: 

1437: \bibitem{Rajagopal:2000wf}

1438:   K.~Rajagopal and F.~Wilczek,

1439:   %``The condensed matter physics of QCD,''

1440:   arXiv:hep-ph/0011333.

1441:   %%CITATION = HEP-PH 0011333;%%

1442: 

1443: \bibitem{alford_review}

1444:   M.~G.~Alford,

1445:   %``Color superconducting quark matter,''

1446:   Ann.\ Rev.\ Nucl.\ Part.\ Sci.\  {\bf 51}, 131 (2001)

1447:   [arXiv:hep-ph/0102047].

1448:   %%CITATION = HEP-PH 0102047;%%

1449: 

1450: \bibitem{Alford:2002ng}

1451: M.~G.~Alford,

1452: %``QCD at high density / temperature,''

1453: Nucl.\ Phys.\ Proc.\ Suppl.\  {\bf 117} (2003) 65

1454: [arXiv:hep-ph/0209287].

1455: %%CITATION = HEP-PH 0209287;%%

1456: 

1457: \bibitem{Nardulli:2002ma}

1458: G.~Nardulli,

1459: %``Effective description of QCD at very high densities,''

1460: Riv.\ Nuovo Cim.\  {\bf 25N3} (2002) 1

1461: [arXiv:hep-ph/0202037].

1462: %%CITATION = HEP-PH 0202037;%%

1463: 

1464: \bibitem{Schafer:2003vz}

1465: T.~Schafer,

1466: %``Quark matter,''

1467: arXiv:hep-ph/0304281.

1468: %%CITATION = HEP-PH 0304281;%%

1469: 

1470: \bibitem{Glendenning} 

1471:   N. K. Glendenning, 

1472:   ``Compact stars: Nuclear physics, particle physics, and general relativity'',

1473:   Springer-Verlag,  New York, USA (1997).

1474:   %\href{http://www.slac.stanford.edu/spires/find/hep/www?irn=4025415}{SPIRES entry}

1475: 

1476: \bibitem{Senger:2005pu}

1477:   P.~Senger,

1478:   %``Strangeness and charm of compressed baryonic matter: The CBM experiment at

1479:   %FAIR,''

1480:   J.\ Phys.\ G {\bf 31}, S1111 (2005).

1481:   %%CITATION = JPHGB,G31,S1111;%%

1482: 

1483: \bibitem{bcs1}

1484: J. Bardeen, L.N. Cooper, J.R. Schrieffer, Phys. Rev. \textbf{106}, 162 (1957)

1485: 

1486: \bibitem{Buballa:2002xv}

1487:   M.~Buballa and M.~Oertel,

1488:   %``Color superconductivity in two- and three-flavor systems at moderate

1489:   %densities,''

1490:   arXiv:hep-ph/0205027.

1491:   %%CITATION = HEP-PH 0205027;%%

1492: 

1493: \bibitem{ARW3}

1494:   M.~G.~Alford, K.~Rajagopal and F.~Wilczek,

1495:   %``Color-flavor locking and chiral symmetry breaking in high density {QCD},''

1496:   Nucl.\ Phys.\ B {\bf 537}, 443 (1999)

1497:   [arXiv:hep-ph/9804403].

1498:   %%CITATION = HEP-PH 9804403;%%

1499: 

1500: \bibitem{Alford:1999pa}

1501: M.~G.~Alford, J.~Berges and K.~Rajagopal,

1502: %``Unlocking color and flavor in superconducting strange quark matter,''

1503: Nucl.\ Phys.\ B {\bf 558} (1999) 219

1504: [arXiv:hep-ph/9903502].

1505: %%CITATION = HEP-PH 9903502;%%

1506: 

1507: \bibitem{Buballa:2001gj}

1508: M.~Buballa and M.~Oertel,

1509: %``Color-flavor unlocking and phase diagram with self-consistently  determined strange quark masses,''

1510: Nucl.\ Phys.\ A {\bf 703} (2002) 770

1511: [arXiv:hep-ph/0109095].

1512: %%CITATION = HEP-PH 0109095;%%

1513: 

1514: \bibitem{IwaIwa}

1515:  %\bibitem{Iwasaki:1994ij}

1516:   M.~Iwasaki and T.~Iwado,

1517:   %``Superconductivity in the quark matter,''

1518:   Phys.\ Lett.\ B {\bf 350}, 163 (1995).

1519:   %%CITATION = PHLTA,B350,163;%%

1520:  %\bibitem{Iwasaki:1995xf}

1521:   M.~Iwasaki and T.~Iwado,

1522:   %``Hartree-Bogolyubov theory of quark matter: Condensation energy in strange

1523:   %quark matter,''

1524:   Prog.\ Theor.\ Phys.\  {\bf 94}, 1073 (1995).

1525:   %%CITATION = PTPKA,94,1073;%%

1526:  %\bibitem{Iwasaki:2000vw}

1527:   M.~Iwasaki, S.~Ishikawa and T.~Tanaka,

1528:   %``Antiparallel spin pairing in color superconductivity,''

1529:   Prog.\ Theor.\ Phys.\  {\bf 104}, 1041 (2000).

1530:   %%CITATION = PTPKA,104,1041;%%

1531: 

1532: \bibitem{TS1flav}

1533:   T.~Schafer,

1534:   %``Quark hadron continuity in QCD with one flavor,''

1535:   Phys.\ Rev.\ D {\bf 62}, 094007 (2000)

1536:   [arXiv:hep-ph/0006034].

1537:   %%CITATION = HEP-PH 0006034;%%

1538: 

1539: \bibitem{BHO}

1540: %\cite{Buballa:2002wy}

1541: M.~Buballa, J.~Hosek and M.~Oertel,

1542: %``Anisotropic admixture in color-superconducting quark matter,''

1543: Phys.\ Rev.\ Lett.\  {\bf 90} (2003) 182002

1544: [arXiv:hep-ph/0204275].

1545: %%CITATION = HEP-PH 0204275;%%

1546: 

1547: \bibitem{Alford:2002kj}

1548: M.~G.~Alford, J.~A.~Bowers, J.~M.~Cheyne and G.~A.~Cowan,

1549: %``Single color and single flavor color superconductivity,''

1550: Phys.\ Rev.\ D {\bf 67} (2003) 054018

1551: [arXiv:hep-ph/0210106].

1552: %%CITATION = HEP-PH 0210106;%%

1553: 

1554: \bibitem{Schmitt:2004hg}

1555:   A.~Schmitt,

1556:   %``Spin-one color superconductivity in cold and dense quark matter,''

1557:   arXiv:nucl-th/0405076.

1558:   %%CITATION = NUCL-TH 0405076;%%

1559: 

1560: \bibitem{Schmitt:2004et}

1561: A.~Schmitt,

1562: %``The ground state in a spin-one color superconductor,''

1563: Phys.\ Rev.\ D {\bf 71} (2005) 054016

1564: [arXiv:nucl-th/0412033].

1565: %%CITATION = NUCL-TH 0412033;%%

1566: 

1567: \bibitem{Alford:2003fq}

1568: M.~Alford, C.~Kouvaris and K.~Rajagopal,

1569: %``Gapless color-flavor-locked quark matter,''

1570: Phys.\ Rev.\ Lett.\  {\bf 92} (2004) 222001

1571: [arXiv:hep-ph/0311286].

1572: %%CITATION = HEP-PH 0311286;%%

1573: 

1574: \bibitem{Alford:2004hz}

1575:   M.~Alford, C.~Kouvaris and K.~Rajagopal,

1576:   %``Evaluating the gapless color-flavor locked phase,''

1577:   Phys.\ Rev.\ D {\bf 71}, 054009 (2005)

1578:   [arXiv:hep-ph/0406137].

1579:   %%CITATION = HEP-PH 0406137;%%

1580: 

1581: \bibitem{OurLOFF}

1582: M.~Alford, J.~Bowers and K.~Rajagopal,

1583: %``Crystalline color superconductivity,''

1584: Phys.\ Rev.\ D {\bf 63}, 074016 (2001)

1585: [hep-ph/0008208].

1586: %%CITATION = HEP-PH 0008208;%%

1587: 

1588: \bibitem{Bowers:2002xr}

1589: J.~A.~Bowers and K.~Rajagopal,

1590: %``The crystallography of color superconductivity,''

1591: Phys. Rev. D {\bf 66}, 065002 (2002) [hep-ph/0204079].

1592: %%CITATION = HEP-PH 0204079;%%

1593: 

1594: \bibitem{Bowers_thesis} 

1595: J.~A.~Bowers,

1596: %``Color superconducting phases of cold dense quark matter,''

1597: arXiv:hep-ph/0305301.

1598: %%CITATION = HEP-PH 0305301;%%

1599: 

1600: 

1601: \bibitem{Ruster:2005jc}

1602:   S.~B.~Ruster, V.~Werth, M.~Buballa, I.~A.~Shovkovy and D.~H.~Rischke,

1603:   %``The phase diagram of neutral quark matter: Self-consistent treatment of

1604:   %quark masses,''

1605:   Phys.\ Rev.\ D {\bf 72}, 034004 (2005)

1606:   [arXiv:hep-ph/0503184].

1607:   %%CITATION = HEP-PH 0503184;%%

1608: 

1609: 

1610: \bibitem{Abuki:2005ms}

1611:   H.~Abuki and T.~Kunihiro,

1612:   %``Extensive study of phase diagram for charge neutral homogeneous quark

1613:   %matter affected by dynamical chiral condensation: Unified picture for thermal

1614:   %unpairing transitions from weak to strong coupling,''

1615:   arXiv:hep-ph/0509172.

1616:   %%CITATION = HEP-PH 0509172;%%

1617: 

1618: \bibitem{Steiner:2002gx}

1619: A.~W.~Steiner, S.~Reddy and M.~Prakash,

1620: %``Color-neutral superconducting quark matter,''

1621: Phys.\ Rev.\ D {\bf 66} (2002) 094007

1622: [arXiv:hep-ph/0205201].

1623: %%CITATION = HEP-PH 0205201;%%

1624: 

1625: \bibitem{Ruster:2005ib}

1626:   S.~B.~Ruster, V.~Werth, M.~Buballa, I.~A.~Shovkovy and D.~H.~Rischke,

1627:   %``The phase diagram of neutral quark matter: The effect of neutrino

1628:   %trapping,''

1629:   arXiv:hep-ph/0509073.

1630:   %%CITATION = HEP-PH 0509073;%%

1631: 

1632: \bibitem{Huang:2001yw}

1633: M.~Huang, P.~f.~Zhuang and W.~q.~Chao,

1634: %``Massive quark propagator and competition between chiral and diquark

1635: %condensate,''

1636: Phys.\ Rev.\ D {\bf 65} (2002) 076012

1637: [arXiv:hep-ph/0112124].

1638: %%CITATION = HEP-PH 0112124;%%

1639: 

1640: \bibitem{Blaschke:2003cv}

1641: D.~Blaschke, M.~K.~Volkov and V.~L.~Yudichev,

1642: %``Coexistence of color superconductivity and chiral symmetry breaking  within the NJL model,''

1643: arXiv:hep-ph/0301065.

1644: %%CITATION = HEP-PH 0301065;%%

1645: 

1646: \bibitem{helium3} 

1647: D.~Vollhardt and P.~W\"{o}lfle, The Superfluid

1648:   Phases of Helium 3, Taylor and Francis, (1990).

1649: 

1650: \bibitem{Gerhold:2003js}

1651: A.~Gerhold and A.~Rebhan,

1652: %``Gauge dependence identities for color superconducting QCD,''

1653: Phys.\ Rev.\ D {\bf 68} (2003) 011502

1654: [arXiv:hep-ph/0305108].

1655: %%CITATION = HEP-PH 0305108;%%

1656: 

1657: 

1658: \bibitem{Dietrich:2003nu}

1659: D.~D.~Dietrich and D.~H.~Rischke,

1660: %``Gluons, tadpoles, and color neutrality in a two-flavor color

1661: %superconductor,''

1662: Prog.\ Part.\ Nucl.\ Phys.\  {\bf 53} (2004) 305

1663: [arXiv:nucl-th/0312044].

1664: %%CITATION = NUCL-TH 0312044;%%

1665: 

1666: \bibitem{Gerhold:2004sk}

1667: A.~Gerhold,

1668: %``Gluonic contributions to the CSC color charge density are negligible,''

1669: arXiv:hep-ph/0411086.

1670: %%CITATION = HEP-PH 0411086;%%

1671: 

1672: \bibitem{Buballa:2005bv}

1673:   M.~Buballa and I.~A.~Shovkovy,

1674:   %``A note on color neutrality in NJL-type models,''

1675:   Phys.\ Rev.\ D {\bf 72}, 097501 (2005)

1676:   [arXiv:hep-ph/0508197].

1677:   %%CITATION = HEP-PH 0508197;%%

1678: 

1679: \bibitem{Bedaque:2001je}

1680: P.~F.~Bedaque and T.~Schafer,

1681: %``High density quark matter under stress,''

1682: Nucl.\ Phys.\ A {\bf 697}, 802 (2002)

1683: [arXiv:hep-ph/0105150];

1684: 

1685: \bibitem{Neumann:2002jm}

1686: F.~Neumann, M.~Buballa and M.~Oertel,

1687: %``Mixed phases of color superconducting quark matter,''

1688: Nucl.\ Phys.\ A {\bf 714} (2003) 481

1689: [arXiv:hep-ph/0210078].

1690: %%CITATION = HEP-PH 0210078;%%

1691: 

1692: \bibitem{Reddy:2004my}

1693:   S.~Reddy and G.~Rupak,

1694:   %``Phase structure of 2-flavor quark matter: Heterogeneous superconductors,''

1695:   Phys.\ Rev.\ C {\bf 71}, 025201 (2005)

1696:   [arXiv:nucl-th/0405054].

1697:   %%CITATION = NUCL-TH 0405054;%%

1698: 

1699: \bibitem{Kikkawa:1976fe}

1700:   K.~Kikkawa,

1701:   %``Quantum Corrections In Superconductor Models,''

1702:   Prog.\ Theor.\ Phys.\  {\bf 56}, 947 (1976).

1703:   %%CITATION = PTPKA,56,947;%%

1704: 

1705: \bibitem{Ebert:1991pz}

1706: D.~Ebert, L.~Kaschluhn and G.~Kastelewicz,

1707: %``Effective meson - diquark Lagrangian and mass formulas from the Nambu-Jona-Lasinio model,''

1708: Phys.\ Lett.\ B {\bf 264} (1991) 420.

1709: %%CITATION = PHLTA,B264,420;%%

1710: 

1711: \bibitem{Reinhardt:1989rw}

1712: H.~Reinhardt,

1713: %``Hadronization Of Quark Flavor Dynamics,''

1714: Phys.\ Lett.\ B {\bf 244} (1990) 316.

1715: %%CITATION = PHLTA,B244,316;%%

1716: 

1717: \bibitem{Berges:1998rc}

1718: J.~Berges and K.~Rajagopal,

1719: %``Color superconductivity and chiral symmetry restoration at nonzero  baryon density and temperature,''

1720: Nucl.\ Phys.\ B {\bf 538} (1999) 215

1721: [arXiv:hep-ph/9804233].

1722: %%CITATION = HEP-PH 9804233;%%

1723: 

1724: \bibitem{Shovkovy:2003uu}

1725: I.~Shovkovy and M.~Huang,

1726: %``Gapless two-flavor color superconductor,''

1727: Phys.\ Lett.\ B {\bf 564} (2003) 205

1728: [arXiv:hep-ph/0302142].

1729: %%CITATION = HEP-PH 0302142;%%

1730: 

1731: \bibitem{Huang:2003xd}

1732: M.~Huang and I.~Shovkovy,

1733: %``Gapless color superconductivity at zero and at finite temperature,''

1734: Nucl.\ Phys.\ A {\bf 729} (2003) 835

1735: [arXiv:hep-ph/0307273].

1736: %%CITATION = HEP-PH 0307273;%%

1737: 

1738: \bibitem{Aguilera:2005tg}

1739:   D.~N.~Aguilera, D.~Blaschke, M.~Buballa and V.~L.~Yudichev,

1740:   %``Color-spin locking phase in two-flavor quark matter for compact star

1741:   %phenomenology,''

1742:   Phys.\ Rev.\ D {\bf 72}, 034008 (2005)

1743:   [arXiv:hep-ph/0503288].

1744:   %%CITATION = HEP-PH 0503288;%%

1745: 

1746: \bibitem{Schmitt:2005wg}

1747:   A.~Schmitt, I.~A.~Shovkovy and Q.~Wang,

1748:   %``Neutrino emission and cooling rates of spin-one color superconductors,''

1749:   arXiv:hep-ph/0510347.

1750:   %%CITATION = HEP-PH 0510347;%%

1751: 

1752: \bibitem{AR-02}

1753: M.~Alford and K.~Rajagopal,

1754: %``Absence of two-flavor color superconductivity in compact stars,''

1755: JHEP {\bf 0206} (2002) 031

1756: [hep-ph/0204001].

1757: %%CITATION = HEP-PH 0204001;%%

1758: 

1759: \bibitem{Schmitt:2002sc}

1760: A.~Schmitt, Q.~Wang and D.~H.~Rischke,

1761:  %``When the transition temperature in color superconductors is not like in BCS

1762: %theory,''

1763: Phys.\ Rev.\ D {\bf 66} (2002) 114010

1764: [arXiv:nucl-th/0209050].

1765: %%CITATION = NUCL-TH 0209050;%%

1766: 

1767: \bibitem{Bowers:2001ip}

1768: J.~A.~Bowers, J.~Kundu, K.~Rajagopal and E.~Shuster,

1769: %``A diagrammatic approach to crystalline color superconductivity,''

1770: Phys.\ Rev.\ D {\bf 64} (2001) 014024

1771: [arXiv:hep-ph/0101067].

1772: %%CITATION = HEP-PH 0101067;%%

1773: 

1774: \bibitem{Blaschke:2005uj}

1775:   D.~Blaschke, S.~Fredriksson, H.~Grigorian, A.~M.~Oztas and F.~Sandin,

1776:   %``The phase diagram of three-flavor quark matter under compact star

1777:   %constraints,''

1778:   Phys.\ Rev.\ D {\bf 72}, 065020 (2005)

1779:   [arXiv:hep-ph/0503194].

1780:   %%CITATION = HEP-PH 0503194;%%

1781: 

1782: \bibitem{Huang:2004am}

1783:   M.~Huang and I.~A.~Shovkovy,

1784:   %``Screening masses in neutral two-flavor color superconductor,''

1785:   Phys.\ Rev.\ D {\bf 70}, 094030 (2004)

1786:   [hep-ph/0408268].

1787:   %%CITATION = HEP-PH 0408268;%%

1788: 

1789: \bibitem{Casalbuoni:2004tb}

1790:   R.~Casalbuoni, R.~Gatto, M.~Mannarelli, G.~Nardulli and M.~Ruggieri,

1791:   %``Meissner masses in the gCFL phase of QCD,''

1792:   Phys.\ Lett.\ B {\bf 605}, 362 (2005)

1793:   [Erratum-ibid.\ B {\bf 615}, 297 (2005)]

1794:   [hep-ph/0410401].

1795:   %%CITATION = HEP-PH 0410401;%%

1796: 

1797: \bibitem{Giannakis:2004pf}

1798:   I.~Giannakis and H.~C.~Ren,

1799:   %``Chromomagnetic instability and the LOFF state in a two flavor color

1800:   %superconductor,''

1801:   Phys.\ Lett.\ B {\bf 611}, 137 (2005)

1802:   [hep-ph/0412015].

1803:   %%CITATION = HEP-PH 0412015;%%

1804: 

1805: \bibitem{Fukushima:2005cm}

1806:   K.~Fukushima,

1807:   %``Analytical and numerical evaluation of the Debye and Meissner masses in

1808:   %dense neutral three-flavor quark matter,''

1809:   hep-ph/0506080.

1810:   %%CITATION = HEP-PH 0506080;%%

1811: 

1812: \bibitem{Schmitt:2003xq}

1813:   A.~Schmitt, Q.~Wang and D.~H.~Rischke,

1814:   %``Electromagnetic Meissner effect in spin-one color superconductors,''

1815:   Phys.\ Rev.\ Lett.\  {\bf 91} (2003) 242301

1816:   [arXiv:nucl-th/0301090].

1817:   %%CITATION = NUCL-TH 0301090;%%

1818: 

1819: \bibitem{Schmitt:2003aa}

1820:   A.~Schmitt, Q.~Wang and D.~H.~Rischke,

1821:   %``Mixing and screening of photons and gluons in a color superconductor,''

1822:   Phys.\ Rev.\ D {\bf 69} (2004) 094017

1823:   [arXiv:nucl-th/0311006].

1824:   %%CITATION = NUCL-TH 0311006;%%

1825: 

1826: \end{thebibliography}

1827: 

1828: \end{document}

1829: 

1830: 

1831: 

1832: