1:
2: % preprint style
3: \documentclass[preprint,aps]{revtex4}
4: %\documentclass{article}
5: \usepackage{graphicx}
6:
7: %useful macros
8: \newcommand{\bea}{\begin{eqnarray}}
9: \newcommand{\eea}{\end{eqnarray}}
10: \newcommand{\be}{\begin{equation}}
11: \newcommand{\ee}{\end{equation}}
12: \newcommand{\Tr}{\mathrm{Tr}}
13:
14: \begin{document}
15: \title{Meson current in the CFL phase}
16: \author{A.~Gerhold and T.~Sch\"afer}
17: \affiliation{Department of Physics\\
18: North Carolina State University,
19: Raleigh, NC 27695}
20:
21: \begin{abstract}
22: We study the stability of the color-flavor locked (CFL) phase
23: of dense quark matter with regard to the formation of a non-zero
24: Goldstone boson current. We show that an instability appears in
25: the vicinity of the point $\mu_s=\Delta$ which marks the appearance
26: of gapless fermion modes in the CFL phase. Here, $\mu_s=m_s^2/(2\mu)$
27: is the shift in chemical potential due to the strange quark mass
28: and $\Delta$ is the gap in the chiral limit. We show that in
29: the Goldstone boson current phase all components of the magnetic
30: screening mass are real. In this work we do not take into account
31: homogeneous kaon condensation. We study the effects of an instanton
32: induced interaction of the magnitude required to suppress kaon
33: condensation.
34:
35: \end{abstract}
36: \maketitle
37:
38: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
39: \section{Introduction}
40: \label{sec_intro}
41: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
42:
43:
44: Exploring the phase diagram of dense baryonic matter is an area of
45: significant theoretical and experimental activity. Using methods based
46: on weak-coupling QCD it was shown that the groundstate of three flavor
47: quark matter at very high baryon density is the color-flavor-locked (CFL)
48: phase \cite{Alford:1999mk,Schafer:1999fe,Evans:1999at}. Finding the
49: second, third, etc.~most dense phase turns out be a much harder
50: problem. If the quarks are degenerate and flavor symmetry is exact
51: then the symmetries of the CFL state match the symmetries of (hyper)
52: nuclear matter. This means that high density quark matter and low
53: density nuclear matter might be continuously connected, without an
54: intervening phase transition \cite{Schafer:1998ef}.
55:
56: In the real world the strange quark mass is not equal to the masses
57: of the up and down quark and flavor symmetry is broken. At high baryon
58: density the effect of the quark masses is governed by the shift
59: $\mu_q = m_q^2/(2\mu)$ of the Fermi energy due to the quark mass.
60: This implies that flavor symmetry breaking becomes more important as
61: the density is lowered. Clearly, the most important effect is due
62: to the strange quark mass. There are two scales that are important
63: in relation to $\mu_s$. The first is the mass of the lightest
64: strange Goldstone boson, the kaon, and the second is the gap $\Delta$
65: for fermionic excitations. When $\mu_s$ becomes equal to $m_K$
66: the CFL phase undergoes a transition to a phase with kaon condensation
67: \cite{Bedaque:2001je}. In the present work we wish to study possible
68: phase transitions in the vicinity of the point $\mu_s=\Delta$.
69:
70: Alford, Kouvaris and Rajagopal observed that gapless fermions appear
71: in the spectrum if $\mu_s>\Delta$ and that the CFL state is energetically
72: stable compared to other homogeneous phases as long as $\mu_s<2\Delta$.
73: They termed the phase in the regime $\Delta<\mu_s<2\Delta$ the gapless CFL
74: (gCFL) phase \cite{Alford:2003fq,Alford:2004hz}. The problem is that gapless
75: fermion modes in weakly coupled superfluids tend to cause instabilities
76: in current-current correlation functions \cite{Wu:2003,Huang:2004bg}
77: and these instabilities have indeed been found in the gCFL phase
78: \cite{Casalbuoni:2004tb,Alford:2005qw,Fukushima:2005cm}. We have
79: suggested that the instability is resolved by the formation of a
80: non-zero Goldstone boson current \cite{Kryjevski:2005qq,Schafer:2005ym}.
81: We will refer to this phase as the current-CFL (curCFL) phase.
82:
83: In the present work we wish to show that the spacelike screening
84: (Meissner) masses in the curCFL phase are real, and that the
85: instability is indeed resolved. We also improve on the calculation
86: of \cite{Kryjevski:2005qq,Schafer:2005ym} in several ways. We
87: allow the CFL gaps $\Delta_{1,2,3}$ to be different and we properly
88: implement electric charge neutrality. On the other hand, we make
89: one important simplification as compared to our earlier work and
90: ignore homogeneous Goldstone boson condensates. This assumption
91: simplifies the calculation of the dispersion relations and allows
92: a more detailed comparison with previous work on the gCFL phase.
93: We will comment on the effect of kaon condensation in the conclusions.
94: We also compare our results to studies of the LOFF phase in three
95: flavor QCD \cite{Alford:2000ze,Casalbuoni:2003wh,Casalbuoni:2005zp,Ciminale:2006sm,Mannarelli:2006fy}.
96: Recent work on Goldstone boson currents in two flavor QCD can be
97: found in \cite{Giannakis:2004pf,Huang:2005pv,Hong:2005jv,Gorbar:2006up}.
98:
99: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
100: \section{Effective Lagrangian and dispersion laws}
101: \label{sec_disp}
102: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
103:
104: We consider an effective lagrangian that describes the interaction
105: of gapped fermions with background gauge fields
106: \cite{Kryjevski:2004jw,Kryjevski:2004kt,Kryjevski:2005qq,Schafer:2005ym}
107: \bea
108: \label{l_eff}
109: {\cal L } &=&
110: \Tr\left(\chi_L^\dagger(iv\cdot\partial -\hat{\mu}^L-A_e Q)\chi_L\right)
111: + \Tr\left(\chi_R^\dagger(iv\cdot\partial -\hat{\mu}^R-A_e Q)\chi_R\right)
112: \nonumber \\
113: & & \mbox{}
114: -i \Tr \left(\chi_L^\dagger \chi_L X v\cdot(\partial-iA^T)X^\dagger\right)
115: -i \Tr \left(\chi_R^\dagger \chi_R Y v\cdot(\partial-iA^T)Y^\dagger\right)
116: \nonumber \\
117: & & \mbox{}
118: -\frac{1}{2}\sum_{a,b,i,j,k}
119: \Delta_k\left(\chi_{L}^{ai}\chi_{L}^{bj}
120: \epsilon_{kab}\epsilon_{kij}
121: -\chi_{R}^{ai}\chi_{R}^{bj}
122: \epsilon_{kab}\epsilon_{kij} + h.c. \right).
123: \eea
124: Here, $\chi_{L,R}^{ai}$ are left/right handed fermions with color
125: index $a$ and flavor index $i$, $A_\mu$ are $SU(3)_C$ color gauge
126: fields, and $\hat{\mu}^L=MM^\dagger/(2\mu)$, $\hat{\mu}^R=M^\dagger
127: M/(2\mu)$ are effective chemical potentials induced by the quark
128: mass matrix $M$. The matrix $Q=\mathrm{diag}({2\over3},-{1\over3},
129: -{1\over3})$ is the quark charge matrix and $A_e$ is an electro-static
130: potential. The fields $X,Y$ determine the flavor orientation of the
131: left and right handed gap terms and transform as $X\to LXC^T$, $Y\to
132: RYC^T$ under $(L,R)\in SU(3)_L \times SU(3)_R$ and $C\in SU(3)_C$, and
133: $\Delta_k$ $(k=1,2,3)$ are the CFL gap parameters. The gauge field
134: $A_\mu$ appears transposed because of the transformation properties
135: of $X,Y$. From the lagrangian equ.~(\ref{l_eff}) we can read off the
136: Nambu-Gor'kov propagator (left handed part)
137: \be
138: \label{prop}
139: \left(
140: \begin{array}{cc}
141: G^+ & \Xi^- \\
142: \Xi^+ & G^- \end{array}
143: \right)=\left(
144: \begin{array}{cc}
145: (p_0-p){\bf1}+\mathcal{X}_{v}& \underline{\Delta} \\
146: \underline{\Delta} & (p_0+p){\bf1}- \mathcal{X}^T_{-v}
147: \end{array}
148: \right)^{-1},
149: \ee
150: where $p=\vec v\cdot\vec p-\mu$, with the Fermi velocity $\vec v$.
151: The components of the propagator are matrices in color-flavor
152: space. We use a basis spanned by the Gell-Mann matrices $\lambda^A$
153: $(A=1,\ldots,8)$ and $\lambda_0=\sqrt{2\over3}{\bf 1}$. In this basis
154: \be
155: \underline{\Delta}^{AB}=-\textstyle{1\over2}\Delta_{ab}\varepsilon_{ija}
156: \varepsilon_{rsb}\lambda^A_{ir} \lambda^B_{js},
157: \qquad \Delta_{ab}=\mathrm{diag}(\Delta_1,\Delta_2,\Delta_3)_{ab}.
158: \ee
159: We will assume that $X=Y=1$ which excludes the possibility of kaon
160: condensation. This assumption significantly simplifies the calculation
161: of the fermion dispersion relations and the current correlation
162: functions. It is possible to suppress kaon condensation by including
163: a large instanton induced interaction \cite{Schafer:2002ty}. We
164: will discuss instanton effects in Sect.~\ref{secinst}. The gauge
165: field vertex in the phase with $X=Y=1$ is
166: \be
167: \mathcal{X}_{v,AB}={\textstyle{1\over2}}
168: \Tr\left[\lambda_A(\hat\mu^L+A_e Q)\lambda_B+
169: \lambda_A\lambda_B v\cdot A^T\right],
170: \ee
171: We define
172: \be
173: \label{a0}
174: A^{0T} = -\hat\mu^L-A_e Q+\tilde A_3\lambda_3+\tilde A_8\lambda_8,
175: \ee
176: where $\tilde A_3$, $\tilde A_8$ are the diagonal components of the
177: color gauge field relative to the leading order solution of the color
178: neutrality condition. We wish to study the possibility of forming
179: a Goldstone boson current. Gauge invariance implies that the free
180: energy only depends on the combinations $\vec{J}_L=X^\dagger(\vec{\nabla}
181: -i\vec{A}^T)X$ and $\vec{J}_R=Y^\dagger(\vec{\nabla}-i\vec{A}^T)Y$. We
182: will restrict ourselves to diagonal currents $\vec{J}_{L,R}$. Within
183: our approximations the free energies of the vector and axial-vector
184: currents $\vec{J}_L=\pm\vec{J}_R$ are degenerate. We will consider
185: the pure vector current $\vec{J}_L=\vec{J}_R=\vec{A}^T$ with
186: \be
187: \label{cur}
188: \vec{A}^{T}={\textstyle{1\over2}}\vec{\jmath}\,
189: \left(\lambda_3+{\textstyle{1\over\sqrt{3}}}\lambda_8
190: -{\textstyle{1\over\sqrt{6}}}\lambda_0\right).
191: \ee
192: This ansatz has the feature that it does not shift the energy of electrically
193: charged fermion modes. We will see that an ansatz of this type is favored
194: if electric neutrality is enforced. The electric charge due to a gapless
195: charged fermion is $\delta Q\sim \mu^2\jmath$. The density of electrons
196: is $n_e\sim \mu_e^3$ and electric neutrality gives $\mu_e\sim (\mu^2
197: \jmath)^{1/3}\gg \jmath$. This means that any solution with $\mu_e
198: \sim \jmath\sim \Delta$ has to have a very small density of gapless
199: charged fermions. We will study a more general ansatz for the current
200: in Sec.~\ref{secgen} and show that the energetically preferred solution
201: is indeed close to the ansatz equ.~(\ref{cur}). From the propagator
202: (\ref{prop}) we obtain the following dispersion laws,
203: \bea
204: \epsilon_1&=&{\tilde A_3\over2}-{\sqrt{3}\tilde A_8\over 2}-\mu_s
205: -{\vec v\cdot\vec \jmath\over2}
206: +\sqrt{\left(p+{\tilde A_3\over2}+{\tilde A_8\over 2\sqrt{3}}\right)^2
207: +\Delta_1^2},
208: \nonumber\\
209: \epsilon_2&=&-{\tilde A_3\over2}+{\sqrt{3}\tilde A_8\over 2}+\mu_s
210: +{\vec v\cdot\vec \jmath\over2}
211: +\sqrt{\left(p+{\tilde A_3\over2}+{\tilde A_8\over 2\sqrt{3}}\right)^2
212: +\Delta_1^2},
213: \nonumber\\
214: \epsilon_3&=&-{\tilde A_3\over2}-{\sqrt{3}\tilde A_8\over 2}+A_e-\mu_s
215: +\sqrt{\left(p-{\tilde A_3\over2}+{\tilde A_8\over 2\sqrt{3}}
216: -{\vec v\cdot\vec \jmath\over2}\right)^2+\Delta_2^2},
217: \nonumber\\
218: \epsilon_4&=&{\tilde A_3\over2}+{\sqrt{3}\tilde A_8\over 2}-A_e+\mu_s
219: +\sqrt{\left(p-{\tilde A_3\over2}+{\tilde A_8\over 2\sqrt{3}}
220: -{\vec v\cdot\vec \jmath\over2}\right)^2+\Delta_2^2},
221: \nonumber\\
222: \epsilon_5&=&\tilde A_3-A_e + \sqrt{\left(p-{\tilde A_8\over \sqrt{3}}
223: -{\vec v\cdot\vec \jmath\over2}\right)^2+\Delta_3^2},
224: \nonumber\\
225: \epsilon_6&=&-\tilde A_3+A_e +\sqrt{\left(p-{\tilde A_8\over \sqrt{3}}
226: -{\vec v\cdot\vec \jmath\over2}\right)^2+\Delta_3^2},
227: \nonumber\\
228: \epsilon_7&=&\sqrt{p^2+\Delta_1^2}+\ldots,\nonumber\\
229: \epsilon_8&=&\sqrt{p^2+\Delta_1^2}+\ldots,\nonumber\\
230: \epsilon_9&=&\sqrt{p^2+4\Delta_1^2}+\ldots.
231: \eea
232: The dots in $\epsilon_{7,8,9}$ denote
233: terms of the order $a^n$ ($n\ge1$) with $a\in\{\tilde
234: A_3, \tilde A_8, \vec v\cdot\vec \jmath, \Delta_2-\Delta_1, \Delta_3-
235: \Delta_1\}$. We observe that without a current $\epsilon_1$ and
236: $\epsilon_3$ become gapless at $\mu_s=\Delta$. We also note that
237: the current shifts the energy of the electrically neutral modes
238: $\epsilon_{1,2}$ and the momenta of the charged modes $\epsilon_{3-6}$.
239:
240:
241: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
242: \section{Free energy}
243: \label{sec_f}
244: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
245:
246: In order to determine the groundstate we need to supplement the
247: effective lagrangian with a potential for the order parameter $\Delta_k$.
248: In QCD this potential is determined by gauge field dynamics. For the
249: present purpose we are not interested in the details of this interaction
250: and we will model the effective potential as a quadratic function
251: of $\Delta_k$. The free energy is given by
252: \be
253: \label{om1}
254: \Omega={1\over G}(\Delta_1^2+\Delta_2^2+\Delta_3^2)
255: -{1\over G_i} \sqrt{\mu_s}\Delta_3^2
256: +{3\mu^2 \jmath^2\over8\pi^2}-{\mu^2\over4\pi^2}
257: \int dp\int dt
258: \sum_{i=1}^9(|\epsilon_i|-|p|),
259: \ee
260: with $t=\cos\theta$. Here, $G$ fixes the magnitude of the gap
261: in the chiral limit and $G_i$ takes into account instanton effects.
262: The structure of the instanton term is determined by vacuum
263: expectation value of the 't Hooft interaction. The value of $G_i$
264: can be computed in perturbative QCD \cite{Schafer:2002ty} but in
265: the present work we will treat it as a free parameter. We first
266: study the case that there are no instanton effects, $1/G_i=0$. The
267: integral in equ.~(\ref{om1}) is quite complicated. We assume that
268: the external fields and currents are small and expand in a set of
269: parameters $b\in\{\tilde A_3/\Delta_1, \tilde A_8/\Delta_1, \jmath/\Delta_1,
270: (\Delta_2-\Delta_1)/\Delta_1, (\Delta_3-\Delta_1)/\Delta_1, (\mu_s-\Delta_1)
271: /\Delta_1\}$. We will study the convergence properties of this expansion
272: below. To second order in $b$ we obtain
273: \bea
274: \label{om2}
275: \Omega & \simeq &
276: {1\over G}\left(\Delta_1^2+\Delta_2^2+\Delta_3^2\right)
277: +{\mu^2\over54\pi^2}
278: \Big[-54\left(\Delta_1^2+\Delta_2^2+\Delta_3^2\right)
279: \log\left({\mu\over\Delta_1}\right)
280: \nonumber\\
281: &&\mbox{}
282: + \Delta_1^2 (15-34\log2)
283: + 2(\Delta_2^2+\Delta_3^2)(21-17\log2)
284: \nonumber\\
285: &&\mbox{}
286: - 16\Delta_1\left(\Delta_2+\Delta_3\right)
287: (6-\log2)+4\Delta_2\Delta_3(3+4\log2)\Big]
288: \nonumber\\
289: &&\mbox{}
290: +{\mu^2\jmath^2\over648\pi^2}(219-32\log2)
291: +{2\mu^2\over9\pi^2}\left(\tilde A_3^2+\tilde A_8^2\right)(-3+2\log2)
292: \nonumber\\
293: &&\mbox{}
294: +H\left(\Delta_1,-{\tilde A_3\over2}
295: -{\tilde A_8\over2\sqrt{3}},-{\tilde A_3\over2}
296: +{\sqrt{3}\tilde A_8\over2}+\mu_s,0,\jmath\right)
297: \nonumber\\
298: &&\mbox{}
299: +H\left(\Delta_2,{\tilde A_3\over2}
300: -{\tilde A_8\over2\sqrt{3}},{\tilde A_3\over2}
301: +{\sqrt{3}\tilde A_8\over2}-A_e+\mu_s,\jmath,0\right)
302: \nonumber\\
303: &&\mbox{}
304: +H\left(\Delta_3,{\tilde A_8\over\sqrt{3}},0,\jmath,0\right),
305: \eea
306: where the function $H$ is given by
307: \bea
308: \label{om3}
309: H(\Delta,a_1,a_2,\jmath_1,\jmath_2)
310: &=& -{\mu^2\over12\pi^2}
311: \left[6\Delta^2\left(1+2\log\left({2\mu\over\Delta}\right)\right)
312: + 12a_1^2+\jmath_1^2\right]
313: \nonumber\\
314: & & \hspace{0.23cm}\mbox{}
315: +{2\mu^2\sqrt{\Delta}\over15\pi^2\jmath_2}
316: \Big[(2a_2-2\Delta-\jmath_2)^{5/2}\Theta(2a_2-2\Delta-\jmath_2)
317: \nonumber\\
318: & & \hspace{1.65cm}\mbox{}
319: -(2a_2-2\Delta+\jmath_2)^{5/2}\Theta(2a_2-2\Delta+\jmath_2)\Big].
320: \eea
321: The gap equations and neutrality equations can be summarized as $E_i=0$,
322: where
323: \bea
324: E_i&=&\frac{\partial\Omega}{\partial\Delta_i}, \hspace{0.25cm}
325: (i=1,2,3) \\
326: E_4&=&\frac{\partial\Omega}{\partial A_e}, \hspace{0.5cm}
327: E_5 = \frac{\partial\Omega}{\partial\tilde A_3},\hspace{0.25cm}
328: E_6 = \frac{\partial\Omega}{\partial\tilde A_8}.
329: \eea
330: From the linear combination $E_4+E_5+{1\over\sqrt{3}}E_6$ we obtain
331: the relation
332: \be
333: \tilde A_8=-\sqrt{3}\tilde A_3
334: \ee
335: Including the $(-A_e^4)$ term in $\Omega$, we find from $E_4$
336: (for $\mu\gg A_e$)
337: \be
338: A_e=\mathrm{max}(0,\mu_s-\Delta_2-\tilde A_3).
339: \ee
340: We assume $\mu_s-\Delta_2-\tilde A_3<0$ (and therefore $A_e=0$), which
341: will turn out to be consistent for the values of $\mu_s$ under consideration.
342: Then we find $\Delta_3=\Delta_2$. Furthermore, $E_2$ is then a linear
343: equation in $\Delta_2$, with the solution
344: \be
345: \Delta_2=\Delta_1\left(1-{54y\over51+54y-4\log2}\right),
346: \ee
347: where $y=\log(\Delta_1/\Delta(0))$ and $\Delta(0)$ is the value of the
348: gap in the chiral limit.
349: The linear combination $E_1-E_4-2E_5$
350: gives a linear equation for $\tilde A_3$, with the solution
351: \begin{equation}
352: \tilde A_3=-\Delta_1{54y\over21-8\log2}
353: \left(1-{36(1+y)\over51+54y-4\log2}\right).
354: \end{equation}
355: There remains one equation (proportional to $E_4+2E_5$),
356: \begin{equation}
357: {\mu^2(21-8\log 2)\over18\pi^2}\tilde A_3
358: -{\mu^2\sqrt{\Delta_1}\over 6\pi^2\jmath}
359: \left[\Theta(s_+)s_+^{3/2}-\Theta(s_-)s_-^{3/2}\right]=0, \label{e14}
360: \end{equation}
361: where $s_\pm=2\mu_s-2\Delta_1-4\tilde A_3\pm \jmath$. We solve equ. (\ref{e14})
362: numerically for $\Delta_1$ for given values of $\mu_s$ and $\jmath$.
363:
364: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
365: \begin{figure}
366: \includegraphics[width=7cm]{omj0a.eps}
367: \includegraphics[width=8cm]{omja.eps}
368: \caption{Left panel: $\Omega/(\mu^2\Delta(0)^2)$ as a function of $\jmath/
369: \Delta(0)$ for $\Delta_1=\Delta_2=\Delta_3=\Delta(0)$, $A_e=\tilde A_3=
370: \tilde A_8=0$ at $\mu_s=0.9683\Delta(0)$ ($O(b^2)$ approximation). Right
371: panel: $\Omega/(\mu^2\Delta(0)^2)$ as a function of $\jmath/\Delta(0)$ with
372: gap equations and neutrality conditions taken into account. The different
373: curves correspond to different orders in the expansion in $b$: second order
374: (black), third order (green), fourth order (blue), fifth order (red). All
375: curves are shown for the respective values of $\mu_{s,crit}$.
376: \label{figomj}}
377: \end{figure}
378: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
379:
380: We observe that $\Omega/(\mu^2\Delta(0)^2)$ is a function of $b/\Delta(0)$
381: and $\Delta_1/\Delta(0)$. This means that in order to study the effective
382: potential as a function of dimensionless variables we do not have to
383: specify the value of $\mu$ and $G$. We first study the dependence of
384: the thermodynamic potential $\Omega$ on the current $\jmath$ for different
385: $\mu_s$. We begin with the simplified case $\Delta_1=\Delta_2=
386: \Delta_3=\Delta(0)$, $A_e=\tilde A_3=\tilde A_8=0$. We find that $\Omega(
387: \jmath)$ develops a nontrivial minimum for $\mu_s>0.9683\Delta(0)$, as
388: shown in the left panel of Fig.~\ref{figomj}. If we take into account the
389: gap equations and neutrality conditions, a nontrivial minimum appears
390: for a slightly larger value of $\mu_s$, namely $\mu_s>\mu_{s,crit}\equiv
391: 0.9916\Delta(0)$, as shown in the right panel of Fig.~\ref{figomj} (black
392: curve). We note that the expectation value of $\jmath$ is considerably
393: smaller than in the simplified case.
394:
395: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
396: \begin{figure}
397: \includegraphics[width=7cm]{jmusa.eps}
398: \caption{Optimum current $\jmath/\Delta(0)$ as a function of $\mu_s/\Delta
399: (0)$. We compare results at different orders in the small parameters $b$
400: (curves labeled as in Fig.~\ref{figomj}).
401: \label{figjmus}}
402: \end{figure}
403: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
404:
405: We have also studied the importance of higher order terms in $b$
406: in the free energy. These terms are not included in
407: equ.~(\ref{om2},\ref{om3}) but they can easily be generated using
408: a symbolic algebra package such as Mathematica or Maple. In
409: Fig.~\ref{figomj} we compare the $O(b^2)$ result to the $O(b^n)$
410: approximation with $n=3,4,5$. At the minimum of $\Omega$ we find
411: $\Delta_3=\Delta_2$ and $A_e=0$ as before. We also see that while
412: the second order approximation gives results that are only qualitatively
413: correct, the differences between the results of the higher order
414: approximations are very small.
415:
416:
417: We may also determine the minimum of the free energy for $\mu_s$ greater than
418: the critical value. Fig.~\ref{figjmus} shows the expectation value of $\jmath$
419: as a function of $\mu_s$. Again, we compare results at different order in
420: the small parameter $b$. We can see that the differences between the results
421: of the higher order approximations are very small even for (somewhat) higher
422: values of $\mu_s$. The highest approximation we discuss is the fifth order
423: approximation. In that case we find that the critical values of $\mu_s$ is
424: given by $\mu_{s,crit}=0.9919 \Delta(0)$. The expectation value of the
425: $\jmath$ is equal to $0.1541 \Delta(0)$ at this point.
426:
427: In Fig.~\ref{figadjmus} we show the gauge potential $\tilde A_3$ and the
428: flavor asymmetries in the gaps, $\Delta_1-\Delta(0)$, $\Delta_2-\Delta(0)$,
429: as a function of $\mu_s$ calculated to $O(b^5)$. We remark that $\epsilon_1$
430: becomes a gapless mode for $\mu_s>\mu_{s,crit}$, while $\epsilon_3$ (the
431: almost quadratic mode of Ref.~\cite{Alford:2004hz}) has at least a very
432: small gap, greater than $0.04 \Delta(0)$, for all values of $\mu_s$.
433:
434: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
435: \begin{figure}
436: \includegraphics[width=8cm]{adjmusa.eps}
437: \caption{Gauge field $\tilde A_3/\Delta(0)$ (dotted) and flavor
438: symmetry breaking in the gap parameters $(\Delta_1-\Delta(0))/\Delta(0)$
439: (continuous) and $(\Delta_2-\Delta(0))/\Delta(0)$ (dashed) as functions
440: of $\mu_s/\Delta(0)$, computed to $O(b^5)$.\label{figadjmus}}
441: \end{figure}
442: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
443:
444:
445:
446: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
447: \section{Meissner masses}
448: \label{sec_meiss}
449: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
450:
451: In this section we study the stability of the current-current
452: correlation functions. The screening (Meissner) masses are given by
453: \bea
454: (m^2_M)_{ab}^{ij} &=& {g^2\mu^2\over2\pi^2} \delta_{ab}\delta^{ij}
455: + {g^2\mu^2 \over16\pi^2}\lim_{k\to0}\lim_{k_0\to0}
456: \int dp\int dt\int {dp_0\over{2\pi}}
457: \nonumber \\[0.2cm]
458: & & \hspace{2cm}\mbox{} \times\mathrm{Tr}
459: \Big[ G^+(p) V_a^i G^+(p+k)V_b^j
460: + G^-(p)\tilde V_a^i G^-(p+k)\tilde V_b^j \nonumber \\
461: & & \hspace{2.7cm}\mbox{}
462: + \Xi^+(p) V_a^i\Xi^-(p+k)\tilde V_b^j
463: + \Xi^-(p)\tilde V_a^i \Xi^+(p+k) V_b^j\Big],
464: \eea
465: where we have defined the vertices
466: \begin{equation}
467: (V_a^i)_{AB}=\textstyle{1\over2}\Tr[\lambda_A\lambda_B\lambda_a^T]v^i,
468: \qquad (\tilde V_a^i)_{AB}=-\textstyle{1\over2}
469: \Tr[\lambda_B\lambda_A\lambda_a^T] \tilde v^i,
470: \end{equation}
471: with $\tilde v^i=-v^i$.
472: We begin by studying the screening masses for small $\mu_s<\mu_{s,crit}$.
473: In this case the Meissner mass matrix is diagonal, and we have $m_{M11}^2
474: =m_{M22}^2=m_{M33}^2=m_{M88}^2$ and $m_{M44}^2=m_{M55}^2
475: =m_{M66}^2=m_{M77}^2$, with
476: \bea
477: m_{M11}^2 &=& g^2\mu^2 \, {21-8\log2\over108\pi^2},\\
478: m_{M44}^2 &=& {g^2\mu^2\over\pi^2}
479: \bigg[-{\log2-3\alpha^2\over6\alpha^2}
480: -{1\over\alpha\beta}\arctan\left({\alpha\over\beta}\right)
481: +{1-\alpha^2\over2\alpha^2\gamma}\mathrm{arctanh}
482: \left({\gamma\over1+\beta^2}\right)\bigg],
483: \eea
484: with $\alpha=\mu_s/\Delta(0)$, $\beta=\sqrt{4-\alpha^2}$ and $\gamma=
485: \sqrt{9-10\alpha^2+\alpha^4}$. We note that the Meissner masses of
486: the modes $i=1,2,3,8$ are independent of $\mu_s$ below $\mu_{s,crit}$.
487: These modes only couple to pairs of quasi-particles that are subject
488: to the same shift in energy. In our approximation, in which we neglect
489: the effect of $\mu_s$ on the density of states, this implies that
490: the $\mu_s $ dependence can be eliminated by a simple shift in the
491: loop energy. This is not the case for the off-diagonal gluon modes
492: $i=4,5,6,7$. We find that the screening masses $m_{Mii}$ ($i=4,5,6,7$)
493: drop with $\mu_s$, but they remain real and non-zero for $\mu_s<
494: \mu_{s,crit}$ \cite{Casalbuoni:2004tb}.
495:
496: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
497: \begin{figure}
498: \includegraphics[width=7cm]{m3a.eps}
499: \includegraphics[width=7cm]{m1a.eps}
500: \caption{Meissner masses squared $m_{M(2)\perp}^2/(g^2\mu^2)$,
501: $m_{M(2)\parallel}^2/(g^2\mu^2)$, $m_{M11\perp}^2/(g^2\mu^2)$ and
502: $m_{M11\parallel}^2/(g^2\mu^2)$ as functions of $\mu_s/\Delta(0)$,
503: computed to $O(b^5)$. \label{figmh}}
504: \end{figure}
505: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
506:
507: In the presence of a finite current $\vec \jmath$ we may decompose the
508: Meissner masses into a longitudinal and a transverse component,
509: \be
510: (m_M^2)^{ij}=m_{M\perp}^2(\delta^{ij}-\hat \jmath^i\hat \jmath^j)
511: +m_{M\parallel}^2\hat \jmath^i\hat \jmath^j.
512: \ee
513: First let us evaluate the Meissner masses through order $b^0$, which
514: corresponds to the $O(b^2)$-approximation in the free energy.
515: We will see that the interesting terms are those of order $b^{-1/2}$.
516:
517: A chromomagnetic instability could occur for the Meissner masses with
518: color indices 1,2,3 or 8 \cite{Casalbuoni:2004tb}. In the 3-8-sector
519: there are contributions where both quark propagators in the gluon self
520: energy diagram are gapless. We find that the mass matrix in the 3-8-sector
521: is not diagonal, the mixing angle being equal to $\pi\over6$. We obtain
522: the following eigenvalues of the 3-8-mass matrix within our approximations,
523: \bea
524: m_{M(1)\perp,\parallel}^2&=&g^2\mu^2{21-8\log2\over108\pi^2},\\
525: m_{M(2)\perp}^2&=&g^2\mu^2\bigg\{{21-8\log2\over108\pi^2}
526: +{\sqrt{\Delta_1}\over45\pi^2\jmath^3}\Big[\Theta(s_+)s_+^{3/2}
527: (2s_+\!-5\jmath) \nonumber\\
528: & & \hspace{0.5cm}\mbox{}
529: -\Theta(s_-)s_-^{3/2}(2s_-\!+5\jmath)\Big] \bigg\},\\
530: m_{M(2)\parallel}^2&=&g^2\mu^2\bigg\{{21-8\log2\over108\pi^2}
531: -{\sqrt{\Delta_1}\over90\pi^2\jmath^3}\big[\Theta(s_+)
532: \sqrt{s_+}(15\jmath^2-20\jmath s_++8s_+^2)\nonumber\\
533: & & \hspace{0.5cm}\mbox{}
534: -\Theta(s_-)\sqrt{s_-}(15\jmath^2+20\jmath s_-+8s_-^2)\big] \bigg\}.
535: \eea
536: These quantities are positive for $\mu_s=\mu_{s,crit}$ when $\jmath$ is
537: chosen such that $\Omega$ is minimal.
538: We remark that the longitudinal Meissner mass squared $m_{M(2)\parallel}^2$
539: is very similar to the second (partial) derivative of $\Omega$ with respect
540: to $\jmath$ (only the $\Theta$-independent terms are different because
541: the current contains a component proportional to $\lambda_0$.)
542:
543: For $m_{M11}^2(=m_{M22}^2)$ there are contributions where one quark propagator
544: in the gluon self energy diagram is gapless and the other one contains the the
545: would-be quadratic mode. We find
546: \bea
547: m_{M11\perp}^2&=&\!g^2\mu^2\Bigg\{{21-8\log2\over108\pi^2}
548: -{\sqrt{\Delta_1}\over60\pi^2\jmath^3}\bigg[\Theta(s_+)
549: \bigg(-\sqrt{s_+}\big(8s_+^2+15r^2+25rs_+\nonumber\\
550: & & \hspace{0.5cm}\mbox{}
551: -10\jmath(2s_++3r)\big)+15\sqrt{r}\left((s_++r-\jmath)^2-\jmath^2\right)
552: \arctan\sqrt{s_+\over r}\bigg)\nonumber\\
553: & & \hspace{0.5cm}\mbox{}
554: -\left(\jmath\leftrightarrow -\jmath,\
555: s_+\leftrightarrow s_-\right)\bigg]\Bigg\}, \\
556: m_{M11\parallel}^2&=&g^2\mu^2\Bigg\{{21-8\log2\over108\pi^2}
557: -{\sqrt{\Delta_1}\over30\pi^2\jmath^3}\bigg[\Theta(s_+)
558: \bigg(\sqrt{s_+}\big(8s_+^2+15r^2+25rs_+\nonumber\\
559: & & \hspace{0.5cm}\mbox{}
560: -5\jmath(4s_++6r-3\jmath)\big)-15
561: \sqrt{r}(s_++r-\jmath)^2\arctan\sqrt{s_+\over r}\bigg)\nonumber\\
562: & & \hspace{0.5cm}\mbox{}
563: -\left(\jmath\leftrightarrow -\jmath,\ s_
564: +\leftrightarrow s_-\right)\bigg]\Bigg\},
565: \eea
566: where $r=2(\Delta_2-\mu_s+\tilde A_3)>0$. Again we find that $m_{M11\perp}^2$
567: and $m_{M11\parallel}^2$ are positive.
568:
569: The squared Meissner masses $m_{M44}^2$ to $m_{M77}^2$ are equal at
570: $O(b^0)$. At the above value of $\mu_s$ we find
571: \be
572: m_{M44}^2={9-\sqrt{3}\pi-3\log2\over18\pi^2},
573: \ee
574: which is again positive.
575:
576: As above we may include higher orders in $b$ in the calculation of the
577: Meissner masses. Fig.~\ref{figmh} shows the ``dangerous'' components of
578: the Meissner masses as functions of $\mu_s$ at $O(b^5)$.
579: We observe that all masses are real.
580: As above the differences between the results of the
581: higher order approximations are very small. The numerical values of
582: the ``dangerous'' components of the Meissner mass just above the phase
583: transition are shown in the first line of Table \ref{tabm}.
584:
585:
586: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
587: \section{An improved ansatz for the current}
588: \label{secgen}
589: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
590:
591: We have studied the effect of using a more general ansatz for the
592: current. Consider
593: \be
594: \vec{A}^{T}={\textstyle{1\over2}}\vec{\jmath}\,\left(c_1\lambda_3
595: +c_2{\textstyle{1\over\sqrt{3}}}\lambda_8
596: -c_3{\textstyle{1\over\sqrt{6}}}\lambda_0\right) \label{jgen}
597: \ee
598: The ansatz given in equ.~(\ref{cur}) corresponds to $c_1=c_2=c_3=1$.
599: We now minimize $\Omega$ with respect to $\jmath$ and the coefficients
600: $c_i$. We can fix $c_1$ such that the explicit $\jmath^2$ term in
601: equ.~(\ref{om1}) remains unchanged as $c_2$ and $c_3$ are varied,
602: which amounts to setting $c_1=\sqrt{(9-2c_2^2-c_3^2)/6}$. At
603: $O(b^5)$ we find $\mu_{s,crit}=0.9918\Delta(0)$, and
604: \be
605: c_1=0.980, \quad c_2=1.000, \quad c_3=1.114. \label{csol}
606: \ee
607: We observe that the exact minimum is indeed quite close to our
608: simple ansatz. The ``dangerous'' components of the Meissner mass
609: are not significantly affected in the vicinity of the phase transition.
610: Their values just above the phase transition are shown in the second
611: line of Table \ref{tabm}.
612:
613: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
614: \begin{table}
615: \begin {tabular}{|c||c|c|c|c|}\hline
616: % & $m_{M(2)\perp}^2$ & $m_{M(2)\parallel}^2$ & $m_{M11\perp}^2$
617: % & $m_{M11\parallel}^2$\\ \hline\hline
618: & $m_{M(2)\perp}^2/m_M^2(0)$ & $m_{M(2)\parallel}^2/m_M^2(0)$
619: & $m_{M11\perp}^2/m_M^2(0)$ & $m_{M11\parallel}^2/m_M^2(0)$\\ \hline\hline
620: $c_i=1$ &
621: % $0.00516$ & $0.00082$ & $0.01025$ & $0.00253$ \\ \hline
622: $0.356$ & $0.056$ & $0.707$ & $0.174$ \\ \hline
623: $c_{1,3}\neq1$ &
624: % $0.00519$ & $0.00089$ & $0.01009$ & $0.00207$ \\ \hline
625: $0.358$ & $0.061$ & $0.696$ & $0.142$ \\ \hline
626: \end{tabular}
627: \caption{Dangerous components of the Meissner masses at $\mu_s=\mu_{s,crit}$
628: in units of the screening mass at $\mu_s=0$ (third order approximation).
629: First line: $c_{1,2,3}=1$ in equ.~(\ref{jgen}), second line: $c_i$
630: according to equ.~(\ref{csol}).
631: \label{tabm}}
632: \end{table}
633: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
634:
635:
636: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
637: \section{Instanton contribution}
638: \label{secinst}
639: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
640:
641: In this section we consider the role of an instanton induced
642: interaction. The value of $G_i$ can be computed in perturbative
643: QCD \cite{Schafer:2002ty} but the result is very sensitive to
644: the QCD scale parameter and to higher order perturbative corrections.
645: In the present work we have neglected kaon condensation and we
646: will therefore explore the effect of an instanton term of the
647: magnitude required to suppress kaon condensation. We have previously
648: argued \cite{Schafer:2002ty} that an instanton terms of this size is
649: problematic from the point of view of instanton phenomenology at zero
650: baryon density, but we will ignore this issue here.
651:
652: In order to suppress kaon condensation, $1/G_i$ has to
653: satisfy
654: \be
655: {1\over G_i}>{f_\pi^2\mu_s^{3/2}\over\Delta_3^2}, \label{c1}
656: \ee
657: where $f_\pi$ is the pion decay constant in the CFL phase. We have
658: used $1/G_i=0.67f_\pi^2\Delta(0)^{-1/2}$ together with the perturbative
659: result for the pion decay constant \cite{Son:1999cm}
660: \be
661: f_\pi^2 = \frac{21-8\log(2)}{18}\left(\frac{\mu^2}{2\pi^2}\right).
662: \ee
663: In this case the effective potential $\Omega(\jmath)$ develops a
664: non-trivial minimum for $\mu_s>\mu_{s,crit}=0.964\Delta(0)$ (using
665: $\Omega$ to $O(b^5)$ and $c_i=1$). At this point we find
666: $\Delta_3=1.21\Delta(0)$, such that the condition in equ.~(\ref{c1})
667: is just met. The magnitude of the current at the onset is $\jmath=
668: 0.17\Delta(0)$ and the shape of the effective potential is not
669: strongly modified by instanton effects. We also find that all
670: Meissner masses are real. The main effect of the instanton term
671: is a non-zero splitting between $\Delta_3$ and $\Delta_{1,2}$ below
672: $\mu_{s,crit}$, as shown in Fig.~\ref{figdeltai}. This effect leads
673: to the small reduction in the critical value of the effective
674: chemical potential, $\mu_{s,crit}=0.964\Delta(0)$ compared to
675: $\mu_{s,crit}=0.992\Delta(0)$ without the instanton term. We
676: also note that because of the larger splitting between the gaps
677: our expansion converges more slowly.
678:
679: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
680: \begin{figure}
681: \includegraphics[width=7cm]{deltaia.eps}
682: \caption{$\Delta_i/\Delta(0)$ as functions of $\mu_s/\Delta(0)$
683: with the instanton contribution taken into account ($1/G_i=0.67
684: f_\pi^2\Delta(0)^{-1/2}$), computed to $O(b^5)$. Below
685: the phase transition $\Delta_1$ and $\Delta_2$ are identical.
686: \label{figdeltai}}
687: \end{figure}
688: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
689:
690: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
691: \section{Conclusions}
692: \label{sec_sum}
693: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
694:
695: We have shown that near the point at which gapless fermion modes
696: appear in the spectrum the CFL phase becomes unstable with respect
697: to the formation of a Goldstone boson current. We have computed
698: the magnetic screening masses in the Goldstone current phase and shown
699: that all masses are real. Our calculation is based on an effective
700: lagrangian of fermions coupled to a mean-field gap term and background
701: gauge potentials. The CFL pairing energy is of order $\mu^2\Delta^2$.
702: We have performed an expansion in powers of $((\mu_s-\Delta)/\Delta,A/\Delta,
703: \jmath/\Delta,\delta\Delta/\Delta)$ and included all orders in this
704: expansion until numerical convergence was obtained. We have neglected
705: all contributions that are suppressed by extra factors of $\Delta/\mu$.
706:
707: In this paper we have not included the effect of a homogeneous kaon
708: condensate. Kaon condensation is likely to play an important role
709: since the kaon condensate modifies the spectrum of fermions in such
710: a way that the lightest mode is charged. We have considered the
711: possibility that the condensate is suppressed by a large instanton
712: term. We find that the instanton term leads to a splitting between
713: the different CFL gaps for $\mu_s<\mu_{s,crit}$ but it does not
714: qualitatively change the effective potential for the Goldstone
715: boson current.
716:
717: The phase studied in this paper is equivalent to a single plane wave
718: LOFF state. The constant gauge field $\vec A$ can be removed by a
719: gauge transformation which leads to the following gap matrix
720: (for $c_i=1$),
721: \be
722: \Delta_{ab}=\mathrm{diag}
723: (\Delta_1 e^{-i \vec x\cdot\vec\jmath},\Delta_2,\Delta_2)_{ab}.
724: \ee
725: This gap matrix is different from the ansatz $\Delta_{ab}=
726: \mathrm{diag}(0,\Delta_2,\Delta_2)_{ab}e^{-i \vec x\cdot\vec\jmath}$
727: considered by Casalbuoni et al.~in \cite{Casalbuoni:2005zp}. One
728: important difference is that we consider the CFL phase near the
729: onset of the Goldstone boson current instability. In that regime
730: all fermions remain paired and the free energy can be computed
731: as an expansion in $(\jmath/\Delta)$. Casalbuoni et al.~consider
732: the CFL phase near the maximum value of $\mu_s$ at which pairing
733: between different flavors is possible and expand the free energy
734: in $(\Delta/\jmath)$. Clearly, it is important to understand how
735: to interpolate between these two extremes. Once a kaon condensate
736: develops the current is no longer diagonal, and it is also not
737: equivalent to a pure gauge field.
738:
739: Acknowledgments: We would like to thank A.~Kryjevski for useful
740: discussions. This work is supported in part by the US Department
741: of Energy grant DE-FG-88ER40388.
742:
743:
744: \begin{thebibliography}{99}
745: \bibitem{Alford:1999mk}
746: M.~Alford, K.~Rajagopal and F.~Wilczek,
747: %``Color-flavor locking and chiral symmetry breaking in high density {QCD},''
748: Nucl.\ Phys.\ {\bf B537}, 443 (1999)
749: [hep-ph/9804403].
750: %%CITATION = HEP-PH 9804403;%%
751:
752: \bibitem{Schafer:1999fe}
753: T.~Sch{\"a}fer,
754: %``Patterns of symmetry breaking in QCD at high baryon density,''
755: Nucl.\ Phys.\ B {\bf 575}, 269 (2000)
756: [hep-ph/9909574].
757: %%CITATION = HEP-PH 9909574;%%
758:
759: \bibitem{Evans:1999at}
760: N.~Evans, J.~Hormuzdiar, S.~D.~Hsu and M.~Schwetz,
761: %``On the QCD ground state at high density,''
762: Nucl.\ Phys.\ B {\bf 581}, 391 (2000)
763: [hep-ph/9910313].
764: %%CITATION = HEP-PH 9910313;%%
765:
766: \bibitem{Schafer:1998ef}
767: T.~Sch{\"a}fer and F.~Wilczek,
768: %``Continuity of quark and hadron matter,''
769: Phys.\ Rev.\ Lett.\ {\bf 82}, 3956 (1999)
770: [hep-ph/9811473].
771: %%CITATION = HEP-PH 9811473;%%
772:
773: \bibitem{Bedaque:2001je}
774: P.~F.~Bedaque and T.~Sch{\"a}fer,
775: %``High density quark matter under stress,''
776: Nucl.\ Phys.\ A {\bf 697}, 802 (2002)
777: [hep-ph/0105150].
778: %%CITATION = HEP-PH 0105150;%%
779:
780: \bibitem{Alford:2003fq}
781: M.~Alford, C.~Kouvaris and K.~Rajagopal,
782: %``Gapless color-flavor-locked quark matter,''
783: Phys.\ Rev.\ Lett.\ {\bf 92}, 222001 (2004)
784: [hep-ph/0311286].
785: %%CITATION = HEP-PH 0311286;%%
786:
787: \bibitem{Alford:2004hz}
788: M.~Alford, C.~Kouvaris and K.~Rajagopal,
789: %``Evaluating the gapless color-flavor locked phase,''
790: Phys.\ Rev.\ D {\bf 71}, 054009 (2005)
791: [hep-ph/0406137].
792: %%CITATION = HEP-PH 0406137;%%
793:
794: \bibitem{Wu:2003}
795: S.-T.~Wu, S.~Yip,
796: Phys.\ Rev.\ A {\bf 67}, 053603 (2003)
797: [cond-mat/0303185].
798:
799: \bibitem{Huang:2004bg}
800: M.~Huang and I.~A.~Shovkovy,
801: %``Chromomagnetic instability in dense quark matter,''
802: Phys.\ Rev.\ D {\bf 70}, 051501 (2004)
803: [hep-ph/0407049].
804: %%CITATION = HEP-PH 0407049;%%
805:
806: \bibitem{Casalbuoni:2004tb}
807: R.~Casalbuoni, R.~Gatto, M.~Mannarelli, G.~Nardulli and M.~Ruggieri,
808: %``Meissner masses in the gCFL phase of QCD,''
809: Phys.\ Lett.\ B {\bf 605}, 362 (2005)
810: [Erratum-ibid.\ B {\bf 615}, 297 (2005)]
811: [hep-ph/0410401].
812: %%CITATION = HEP-PH 0410401;%%
813:
814: \bibitem{Alford:2005qw}
815: M.~Alford and Q.~H.~Wang,
816: %``Photons in gapless color-flavor-locked quark matter,''
817: J.\ Phys.\ G {\bf 31}, 719 (2005)
818: [hep-ph/0501078].
819: %%CITATION = HEP-PH 0501078;%%
820:
821: \bibitem{Fukushima:2005cm}
822: K.~Fukushima,
823: %``Analytical and numerical evaluation of the Debye and Meissner masses in
824: %dense neutral three-flavor quark matter,''
825: Phys.\ Rev.\ D {\bf 72}, 074002 (2005)
826: [hep-ph/0506080].
827: %%CITATION = HEP-PH 0506080;%%
828:
829: \bibitem{Kryjevski:2005qq}
830: A.~Kryjevski,
831: %``Spontaneous superfluid current generation in CFL at nonzero strange quark
832: %mass,''
833: preprint, hep-ph/0508180.
834: %%CITATION = HEP-PH 0508180;%%
835:
836: \bibitem{Schafer:2005ym}
837: T.~Sch\"afer,
838: %``Meson supercurrent state in high density QCD,''
839: Phys.\ Rev.\ Lett.\ {\bf 96}, 012305 (2006)
840: [hep-ph/0508190].
841: %%CITATION = HEP-PH 0508190;%%
842:
843: \bibitem{Alford:2000ze}
844: M.~G.~Alford, J.~A.~Bowers and K.~Rajagopal,
845: %``Crystalline color superconductivity,''
846: Phys.\ Rev.\ D {\bf 63}, 074016 (2001)
847: [hep-ph/0008208].
848: %%CITATION = HEP-PH 0008208;%%
849:
850: \bibitem{Casalbuoni:2003wh}
851: R.~Casalbuoni and G.~Nardulli,
852: %``Inhomogeneous superconductivity in condensed matter and QCD,''
853: Rev.\ Mod.\ Phys.\ {\bf 76}, 263 (2004)
854: [hep-ph/0305069].
855: %%CITATION = HEP-PH 0305069;%%
856:
857: \bibitem{Casalbuoni:2005zp}
858: R.~Casalbuoni, R.~Gatto, N.~Ippolito, G.~Nardulli and M.~Ruggieri,
859: %``Ginzburg-Landau approach to the three flavor LOFF phase of QCD,''
860: Phys.\ Lett.\ B {\bf 627}, 89 (2005)
861: [hep-ph/0507247].
862: %%CITATION = HEP-PH 0507247;%%
863:
864: \bibitem{Ciminale:2006sm}
865: M.~Ciminale, G.~Nardulli, M.~Ruggieri and R.~Gatto,
866: %``Chromomagnetic stability of the three flavor
867: %Larkin-Ovchinnikov-Fulde-Ferrell phase of QCD,''
868: preprint, hep-ph/0602180.
869: %%CITATION = HEP-PH 0602180;%%
870:
871: \bibitem{Mannarelli:2006fy}
872: M.~Mannarelli, K.~Rajagopal and R.~Sharma,
873: %``Testing the Ginzburg-Landau approximation for three-flavor crystalline
874: %color superconductivity,''
875: preprint, hep-ph/0603076.
876: %%CITATION = HEP-PH 0603076;%%
877:
878: \bibitem{Giannakis:2004pf}
879: I.~Giannakis and H.~C.~Ren,
880: %``Chromomagnetic instability and the LOFF state in a two flavor color
881: %superconductor,''
882: Phys.\ Lett.\ B {\bf 611}, 137 (2005)
883: [hep-ph/0412015].
884: %%CITATION = HEP-PH 0412015;%%
885:
886: \bibitem{Huang:2005pv}
887: M.~Huang,
888: %``Spontaneous current generation in the 2SC phase,''
889: Phys.\ Rev.\ D {\bf 73}, 045007 (2006)
890: [hep-ph/0504235].
891: %%CITATION = HEP-PH 0504235;%%
892:
893: \bibitem{Hong:2005jv}
894: D.~K.~Hong,
895: %``RG analysis and magnetic instability in gapless superconductors,''
896: preprint, hep-ph/0506097.
897: %%CITATION = HEP-PH 0506097;%%
898:
899: \bibitem{Gorbar:2006up}
900: E.~V.~Gorbar, M.~Hashimoto, V.~A.~Miransky and I.~A.~Shovkovy,
901: %``Collective excitations, instabilities, and ground state in dense quark
902: %matter,''
903: preprint, hep-ph/0602251.
904: %%CITATION = HEP-PH 0602251;%%
905:
906: \bibitem{Kryjevski:2004jw}
907: A.~Kryjevski and T.~Sch\"afer,
908: %``An effective theory for baryons in the CFL phase,''
909: Phys.\ Lett.\ B {\bf 606}, 52 (2005)
910: [hep-ph/0407329].
911: %%CITATION = HEP-PH 0407329;%%
912:
913: \bibitem{Kryjevski:2004kt}
914: A.~Kryjevski and D.~Yamada,
915: %``CFL phase of high density QCD at non zero strange quark mass,''
916: Phys.\ Rev.\ D {\bf 71}, 014011 (2005)
917: [hep-ph/0407350].
918: %%CITATION = HEP-PH 0407350;%%
919:
920: \bibitem{Schafer:2002ty}
921: T.~Sch\"afer,
922: %``Instanton effects in QCD at high baryon density,''
923: Phys.\ Rev.\ D {\bf 65}, 094033 (2002)
924: [hep-ph/0201189].
925: %%CITATION = HEP-PH 0201189;%%
926:
927: \bibitem{Son:1999cm}
928: D.~T.~Son and M.~A.~Stephanov,
929: %``Inverse meson mass ordering in color-flavor-locking phase of high density
930: %QCD,''
931: Phys.\ Rev.\ D {\bf 61}, 074012 (2000)
932: [hep-ph/9910491].
933: %%CITATION = HEP-PH 9910491;%%
934:
935: \end{thebibliography}
936:
937: \end{document}
938: