1: \documentclass[12pt]{article}

2: \usepackage{epsfig}

3: 

4: \textwidth  16.5cm

5: \textheight 24.5cm

6: \oddsidemargin 0.25cm

7: \parskip .2cm

8: \topmargin -1cm

9: %\renewcommand{\baselinestretch}{1.3}

10: 

11: \begin{document}

12: 

13: \title{Analysis of zero-frequency solutions of the pion dispersion

14: equation in nuclear matter}

15: 

16: \author{V. A. Sadovnikova\\

17: Petersburg Nuclear Physics Institute, \\

18: Gatchina, St.~Petersburg 188300,

19: Russia}

20: \date{}

21: \maketitle

22: \begin{abstract}

23: In this paper we consider instability of nuclear matter which

24: takes place when the frequencies of the collective excitations

25: turn to zero.  We investigate collective excitations with pion

26: quantum numbers $J^\pi=0^-$.  We study the dependence of

27: zero-frequency solutions of the pion dispersion equation on the

28: value of the spin-isospin quasiparticle interaction $G'$. The

29: solutions of the pion dispersion equation describe the different

30: types of the excitations in the matter, $\omega_i(k)$.  At the

31: critical density  $\rho=\rho_c$ one of solutions of the definite

32: type turns to zero: $\omega_{i0}(k_c)$=0. When $\rho>\rho_c$, the

33: excitations $\omega_{i0}(k)$ become amplified.  It is shown that

34: there is such a ``transitional" value of $G'=G'_{tr}$ that for

35: $G'<G'_{tr}$ the zero-frequency solutions belong to the type

36: $\omega_{sd}$ while for $G'>G'_{tr}$ they pertain to the type

37: $\omega_c$. The solutions of the type $\omega_{sd}$ correspond

38: to instability to small density fluctuations of the nuclear

39: matter at $G'\le -1$.  On the other hand,    $\omega_c$ is

40: responsible for the ``pion condensation" at $G'\approx 2$.  For

41: the  stable nuclear matter the branches of solutions

42: $\omega_{sd}(k)$ and $\omega_c(k)$ are located on the unphysical

43: sheets of the complex plane of frequency.

44: \end{abstract}

45: 

46: 

47: 

48: \section{Introduction}

49: 

50:  In this paper we investigate the excitations of nuclear matter

51: with the pion quantum numbers.  Our analysis is based on studies

52: the in-medium pion dispersion equation. Interactions of pions

53: with baryons in nuclear matter is included in framework of Migdal

54: model \cite{Mi}.

55: Solutions of the pion dispersion equation in this model were

56: considered on the physical sheet of the complex plane of pion

57: frequency $\omega$.

58: We expand this analysis to the unphysical sheets of

59: the Riemann surfaces. We have included first the

60: unphysical sheets to the analysis in our earlier papers

61: \cite{DRS,SR}. In \cite{DRS,SR} we studied the branches of

62: solutions responsible for the pion condensation and the long

63: wavelength instability and it was shown that they are the

64: separate branches supplemented to the well-known zero-sound,

65: pion, and isobar ones.

66: 

67: In this paper we continue to study the solutions of zero-sound

68: and pion dispersion equations responsible for the instability of

69: nuclear matter using the retarded pion propagator.  We analyze

70: zero-frequency solutions of the pion dispersion equation

71: depending on the value of $G'$ and define to what type

72: excitations  these zero-frequency solutions refer.  This could

73: permit us to do conclusion about the character of the phase

74: transition related to the considered instability.

75: 

76: 

77: Except the purely theoretical interest the problem has various

78: applications.  Investigation of the pion excitations in  nuclear

79: matter, started long ago, continue to play  an important role in

80: nuclear physics. Special interest in study of the pionic

81: many-body degrees of freedom was stimulated by prediction and

82: investigation of the pion condensation \cite{Mi,SS,BW,OTW}.  In

83: recent years the study of the pionic many-body degrees of

84: freedom is related to the investigations of the excited states

85:  and of phase transitions in nuclear matter in heavy ion

86: collisions.  Detailed knowledge of in-medium pion dynamics is

87: essentially important for description of mesons (\cite{Mi,OTW}),

88: nucleons (\cite{DRS}) and $\Delta$-isobars \cite{MSG},

89: \cite{HR}.  Analysis of the dilepton

90: production from $\pi^+-\pi^-$ annihilation requires the

91: knowledge of the pion dispersion in hot and dense matter \cite

92: {HR,GK,XX}.

93: 

94: To study of the pion dynamics in heavy ion collisions we need

95: the relativistic description of the pion self-energy at the

96: finite temperatures. Solutions of the relativistic pion

97: dispersion equations were presented in papers \cite{LN,AM}. A

98: pion self-energy with the correct nonrelativistic limit was

99: obtained in papers \cite{LP,Lutz}.

100: 

101: When considering  the pion dynamics in nuclear matter at high

102: densities and temperatures, it is important to have as a "point

103: of reference" a reliable description of the nuclear matter

104: excitations with pion quantum numbers in nonrelativistic limit.

105: Such description have been obtained in the pioneering papers of

106: Migdal and his group \cite {Mi}, followed by the numerous papers

107: \cite {BW,OTW,BBN,GOS,CE,DG}.

108: 

109: In this paper we study the solutions of the nonrelativistic pion

110: dispersion equations in symmetric nuclear matter at zero

111: temperature following to \cite{Mi}.

112: The aim of this  paper is to study the solutions

113: with zero frequency $\omega$=0  depending

114: on the value of the coupling constant  $G'$. These solutions

115: characterize the  stability boundary (in this model it is the

116: boundary on the density).  In this paper it is shown that at

117: different values of $G'$ the zero-frequency solutions belong to

118: the different types of excitations.

119: 

120: 

121: The value of $G'$ is considered in the interval $-1\leq G'\leq

122: 2$.  When we change  $G'$, the branches of solutions are changed

123: as well:  certain solutions go over from the physical to

124:  unphysical sheets and vice verse. To identify the solutions on

125: the unphysical sheets it is important to know the solutions on

126: the physical ones.  The branches of solutions which are obtained

127: in the present paper reproduce the results of \cite{Mi,CC,PR} on

128: the physical sheet of the complex plane of $\omega$.  To do

129: comparison with other papers the simple model with the constant

130: effective quasiparticle interaction is very useful.  However,

131: the pion condensation in this model emerges at too low density

132: and this is not consistent with the results of investigations,

133: \cite{Mi}.

134: 

135:  It is well known that the solutions of the pion dispersion

136: equations, $\omega_i(k)$, describe the different types of

137: excitations in the nuclear matter. They are:  spin-isospin

138: zero-sound wave, $\omega_s(k)$, pion wave, $\omega_\pi(k)$,

139: isobar wave, $\omega_\Delta(k)$, and others \cite{Mi,BW,OTW,EW}.

140: The appearance of the solution with $\omega$=0 means that at the

141: certain values of nuclear density $\rho=\rho_c$ and wave vector

142: $k=k_c$ the frequency of a definite type of excitations $i_0$

143:  turns to zero:  $\omega_{i_0}(k_c)$=0.  If $\rho >\rho_c$ one

144:  obtains $Im\omega_{i_0}(k)>0$, and the amplified solution takes

145: place.  This signals the instability of nuclear matter.  The

146: change of $G'$ causes the changes of the

147: values of $\rho_c$ and $k_c$.  Moreover, at a special situation

148: zero-frequency solution passes to another type of excitations

149: $i_1$:  $\omega_{i_1}(k_c)$=0. It was shown in \cite{LaLi} that

150: the phase transition is determined by the type of those

151: excitations which become unstable. Thus it is important to know

152: to what  type of excitations the solutions with $\omega$=0

153: belongs.

154: 

155: In Fig. 1 we show the values of critical densities

156: $\rho_{c}$ and wave vectors

157: $k_{c}$ for which the pion dispersion equation

158: \begin{equation}\label{1}

159:  D^{-1} = \omega^2-k^2-m^2_\pi-\Pi^R(\omega,k)\ =\ 0\ ,

160: \end{equation}

161: has solutions with $\omega$=0 at any value of $G'$.  We are

162: interested in solutions which satisfy  additional restriction:

163: $\left(\frac d{dk}\omega(k)\right)_{k=k_c}$=0.

164: In Eq.~(1) $\Pi^R(\omega,k)$

165: is the pion self-energy part (retarded polarization operator).

166: The pion self-energy is formed on the basic of particle-hole and

167: isobar-hole loops renormalized due to the effective interactions of

168: quasiparticles: $G'_{NN},$ $G'_{N\Delta}$ and $G'_{\Delta\Delta}$.

169: (The effective constant $G'$ is regarded as

170:  $G'_{NN}$ through the paper.) In Fig.~1 the critical density $\rho_c$

171: and wave vector $k_c$ are presented for three models of the pion

172: self-energy. The matter is stable at $\rho<\rho_c$ for the every

173: model.

174: The physical variables which are related

175: to the pion propagator $D$ diverge along the curves in Fig.~1.

176: This corresponds to the phase transition \cite{LaLi} in nuclear

177: matter. At $\rho<\rho_c$, the nuclear matter is considered as a

178: normal Fermi liquid, but at $\rho>\rho_c$ there is   phase

179: transitions into  new state which may include the standing

180: spin-isospin wave or "condensates of excitation".

181: 

182: Let us introduce the useful definitions. The waves

183: $\omega_s(k)$, $\omega_\pi(k)$, $\omega_\Delta(k)$, and others

184: we name branches of solutions. The set of branches of solutions

185: of the same type for the different values of $G'$ composes a

186: family ($\omega_s$, $\omega_\pi$, $\omega_\Delta$, and others).

187: When we say about zero-frequency solutions or $\omega$=0

188: solutions, we mean that the condition

189: $\left(\frac d{dk}\omega(k)\right)_{k=k_c}$=0 is satisfied in

190: this point, as well.

191: 

192: 

193: Return to Fig.~1.

194: There are known two reasons for instability in the framework of

195: Migdal theory \cite{Mi} in the nonrelativistic symmetric cold

196: nuclear matter at not too high densities. In paper

197: \cite{Pom} it was shown that Fermi liquid becomes unstable when

198: the attractions between the quasiparticles is too strong.  In

199: spin-isospin channel this corresponds to $G'<-1$ \cite{Mi}.  In

200: Fig.~1 we see that there are not solutions with $\omega$=0 at

201:  $G'<-1$.  This is because of the matter is unstable at the any

202: density.  Another

203: region of $G'\approx 2$ corresponds to the instability in

204: respect to  transition into a new ground state containing

205: condensate of pions \cite{Mi}.

206: In the present paper the main attention is paid to study that

207: families of solutions which are responsible for instability

208: along the curves in Fig.~1, from long wavelength instability to pion

209: condensation.

210: 

211: We start with the analysis of the zero-sound dispersion equation for

212: spin-isospin excitations with quantum numbers  $J^\pi=0^-$. This

213: equation is closely related to the pion dispersion equation. It was

214: shown in paper \cite{LaLi,PN} that for the case of strong

215: attraction between the quasiparticles ($G'<-1$) the zero-sound

216: dispersion equation has a growing with time solution which

217: corresponds to  instability in response of small-amplitude, long

218: wavelength density fluctuations. For  $G'>-1$ the unstable solutions do

219: not appear \cite{CC,PR}.

220: 

221: In this paper the family of solutions responsible for the long

222: wavelength instability is built. We denote this family as

223: $\omega_{sd}$. The index ``sd" means ``spinodal decomposition".

224: It is known \cite{LaLi}, that the thermodynamic stability

225: condition requires  the partial derivative of the pressure with

226: respect to the volume to be negative, i.e.,

227: $\left(\frac{\partial P}{\partial V}\right)_T < 0$. The process

228: which takes place in matter, when this condition is broken, is

229: known as a spinodal decomposition \cite{MS}. It is shown in

230: \cite{LaLi}, that there is a relation between the effective

231: quasiparticle interaction and the partial derivative

232: $$

233: \frac{\partial P}{\partial V} = -\frac{N}{V^2} \frac{p_F^2}{3m}

234: (1 + F_0),

235: $$

236: $N$ is the number of particles, $F_0$ is the effective scalar-isoscalar

237: quasiparticle interaction. The stability condition is broken at $F_0 <

238: -1$.  For the zero-sound dispersion equation this corresponds to

239: the appearance of the imaginary amplified solutions. It looks reasonable to

240: use the notation $\omega_{sd}(k)$ for such solutions. Here we use the

241: same notation for the analogous solutions, which appear in the

242: spin-isospin channel at $G'<-1$. Moreover, as it is shown below,

243:  at $G' < 0$ all the solutions belong to the same family. We use

244: the notation $\omega_{sd}(k)$ for them as well.

245: The branches $\omega_{sd}(k)$

246: are imaginary and can be found for all values $G' < 0$.

247: 

248: 

249: Now we turn to the pion dispersion equation. In the channel of

250: excitations with the pion quantum numbers the

251: particle-hole interaction can be modified by taking into account

252: the pion-nucleon interactions \cite{Mi}. As a result, collective

253: and particle-hole  excitations are renormalized due to

254: interaction with pions and, vice versa, the pion excitations are

255: changed due to interaction with particle-hole ones in the

256: nuclear matter. It was predicted \cite{Mi,SS}, that the account

257: of the pion-nucleon interaction is responsible for the phase

258: transition into the ground state with the pion condensation.  In

259: Fig.~1 it is demonstrated that due to the pion-nucleon

260: interaction we obtain the phase transition not only for $G' <

261: -1$ but for all (considered) values of $G'$.

262: 

263: For the pion dispersion equation (\ref{1}) we can find the branch

264: of solutions corresponding to  $\omega_{sd}(k)$. We denote it as

265: $\omega^\pi_{sd}(k)$. The correspondence is established in the

266: following way. It will be seen from the form of the pion self-energy

267: that $\omega_{sd}(k)$ are the poles of $\Pi(\omega,k)$, and for

268: each value of  $k$  the frequency $\omega^\pi_{sd}(k)$

269: tends to  $\omega_{sd}(k)$ as

270: pion-nucleon coupling constant $f_{\pi NN}\rightarrow 0$.

271: 

272: We  show also that  there is  a

273: special value of $G'$, denoted by

274: $G_{tr}$ ($|G'_{tr}|\ll 1$). When $G' < G'_{tr}$, the solutions with

275: $\omega$=0 are on the branches $\omega^\pi_{sd}(k)$.

276: When $G'>G'_{tr}$, the type of the family  carrying

277: zero-frequency solutions is changed. We shall refer to $G'_{tr}$ as

278: "transition value".

279: 

280: When $G'>G'_{tr}$, zero-frequency solutions pass to the branches

281: of solutions of another family.  It is denoted as $\omega_c$

282: (``condensate").  The solutions of (1) referred to this family,

283: $\omega_c(k)$, are identified in the following way. In papers

284: \cite{Mi} it was shown that among the solutions of the pion

285: dispersion equation the imaginary solutions emerge under

286: special conditions.  Instability of matter, related to these

287: imaginary solutions  is interpreted as a phase transitions into

288: a new ground state containing the pion condensate. The branch

289: $\omega_c(k)$ reproduces the imaginary solutions presented in

290: \cite{Mi} on the physical sheet of $\omega$.  It is

291: demonstrated that $\omega_c$ is an independent family of

292: solutions along with $\omega_\pi$, $\omega_s$, $\omega_\Delta$

293: or $\omega^\pi_{sd}$ and it is responsible for pion

294: condensation.

295: 

296: In Sect.~2 the expressions for the retarded pion self-energy are

297: presented and the singularities on the complex

298: plane are investigated. We consider the structure of the

299: Riemann surfaces that are defined by the form of the pion self-energy.

300: The unphysical sheets are described.

301: 

302: In Sect.~3 the family $\omega_{sd}$ is considered for

303: the zero-sound and pion dispersion equations. It is shown how

304: zero-frequency solutions   $\omega_{sd}(k_c)$=0 at

305: $\rho=\rho_c$ (Fig.~1) come these branches.

306: 

307: In Sect.~4 the different types of solutions of the pion dispersion

308: equation  are considered on the physical and unphysical sheets.

309: It is shown that there is an independent family of solutions

310: $\omega_c$ that is related to the pion condensation.

311: 

312: In Sect.~5 we determine the value of   $G'_{tr}$  and show how

313: the  solution with $\omega$=0 pass from the family

314: $\omega^\pi_{sd}(k)$  to $\omega_c(k)$ when $G'= G'_{tr}$.

315: 

316: In this paper we solve the dispersion equations  in the symmetric

317: nuclear matter with zero temperature. The equilibrium density is

318: $\rho_0$, $p_{F0}$=0.268~GeV. During the computations, the

319: effective mass of quasiparticles is assumed to be  $m$=0.8$m_0$,

320: while $m_0$=0.94~GeV is the vacuum nucleon mass. The pion mass

321: is $m_\pi$=0.14~GeV, the isobar mass

322: $\Delta=1.23$~GeV (the isobar width is neglected).

323: Following to the paper \cite{Mi}, the effective nucleon quasiparticle

324: interaction can be written as

325: $$

326: {\cal F}(\vec k_1,\vec \sigma_1,\vec

327: \tau_1;\vec k_2,\vec \sigma_2,\vec \tau_2)=C_0( F(\vec k_1,\vec k_2)

328: +F'(\vec k_1,\vec k_2)(\vec \tau_1\vec \tau_2) +G(\vec k_1,\vec

329: k_2)(\vec \sigma_1\vec \sigma_2)

330: $$

331: $$

332: + G'(\vec k_1,\vec k_2)(\vec

333: \sigma_1\vec \sigma_2) (\vec \tau_1\vec\tau_2)),

334: $$

335: where $\vec \tau,

336: \vec \sigma$ are the Pauli matrices in the isospin and spin spaces. The

337: factor $C_0=N_0^{-1}$, where $N_0$ is a density of states for  two

338: sorts of nucleon on the Fermi surfaces, $N_0=2p_Fm/\pi^2$;

339: $F, F', G, G'$ are the Landau-Migdal dimensionless parameters.

340:  In this paper we use

341: a simple model where parameters are taken to be a constant:

342: $G'(\vec k_1,\vec k_2)=G'(p_F,p_F)\equiv G'$.

343: The Landau-Migdal parameter $G'$ found from the analysis of the

344: nuclear experimental data is  $G'\sim 1.5-1.7$ \cite{BBN,

345: SPWW}.  This parameter describes the short-range correlations between

346: the nucleons with a range $\sim \frac1{m_0}$:

347: $V_{corr}=\frac{f^2}{m_\pi^2}g'(\vec \sigma_1\vec \sigma_2)(\vec

348:  \tau_1\vec\tau_2)$. The relation $G'=\frac1{C_0}\frac{f^2}{m_\pi^2}g'$

349: sets the correspondence of the Landau-Migdal parameters to the effective

350: correlative parameters $g'$ \cite{LN,AM,LP}.

351: 

352: The Landau-Migdal

353: parameter $G'=G'_{NN}$ corresponds to rescattering of

354: $NN\leftrightarrow NN$ quasiparticles.  The parameters

355: $G'_{N\Delta}$ and $G'_{\Delta\Delta}$ determine the

356:  rescattering of $N\Delta\leftrightarrow NN$ and

357: $N\Delta\leftrightarrow N\Delta$.

358: The parameters $G'_{N\Delta}$ and $G'_{\Delta\Delta}$

359:  are not changed in the course of

360: paper  and are taken equal to $G'_{N\Delta}$=0.4 and

361: $G'_{\Delta\Delta}$=1.6 \cite{Mi}.

362: 

363: In the paper we take the pion-nucleon coupling constant $f_{\pi

364: NN}=1$ and the pion-nucleon-isobar coupling constant $f_{\pi

365: N\Delta}=2$.

366: 

367: 

368: 

369: 

370: \section{The pion self-energy $\Pi(\omega,k)$}

371: In this Section we present the expressions for the pion self-energy

372: in nuclear matter. These expressions are well known

373: \cite{Mi,EW,PN,FW,DM}. We do an analytical continuation in  $\omega$

374: of the polarization operator $\Pi(\omega,k)$ to

375:  the unphysical sheets of the complex plane.

376: The search of zero-frequency solutions is  related to

377: investigation of the solutions both on the physical and

378: unphysical sheets.

379: 

380: We consider excitations in nuclear matter which are described by the

381: solutions of the pion dispersion equation (1).

382: The causality requirement leads to the known interpretation of the

383: solutions of the dispersion equation in the complex plane of $\omega$.

384: The stable excitations are described by the solutions with real

385: frequencies, placed on the real axis. The damping solutions (resonances)

386: are placed on the unphysical sheets under the cuts, which the

387: polarization operator has on the real axis.

388: This is Landau damping which is related to overlapping of the frequency

389: of excitations with the frequency of free particle-hole pairs (which

390: give the imaginary part in the pion self-energy).  We can introduce the

391: damping of solutions in another way, with the help of the isobar width

392: (taken as the experimental constant, $\Gamma$=115~MeV) or quasiparticle

393: damping constant, for example. In this case the real excitations are

394: displaced to the lower half plane of the physical sheet \cite{DRS,SR}.

395: We do not consider this damping henceforth.

396: 

397: At certain conditions (large density, a

398: strong attraction between particles) the imaginary solutions

399: arise in the upper half-plane of the physical sheet.  They

400: correspond to the excitations with amplify amplitude. Such

401: excitations correspond to an unstable regime. We find the

402: corresponding solutions with negative imaginary part in the

403: lower half-plane of the physical sheet, as well.

404: 

405:  The amplified solutions in the spin-isospin channel

406: (as well as all the solutions with $Re\omega_i>0$)

407: are interpreted, following \cite{Mi,BaW}, as the

408: frequencies of $\pi^+$-type excitations.  While the solutions in the

409: lower half-plane

410: (as well as all the solutions with $Re\omega_i<0$)

411: are considered as $\pi^-$-type excitations. Their frequencies have to

412: be taken with an opposite sign in respect to the solutions.

413:  For the physical excitations of

414: $\pi^0$-type we put into correspondence the solutions with

415: $Re\omega_i>0$.  In symmetrical nuclear matter with $N=Z$ all three

416: types of solutions, i.e., $\pi^+, \pi^-, \pi^0$ are identical.

417: In the asymmetric nuclear matter the

418: solutions carrying  different charges are transformed in a different

419: way with the change of the parameter of asymmetry and of the density.  The

420: method used in this paper permits to follow these transformations

421:  from  symmetric up to  neutron nuclear matter.

422: 

423: 

424: Pion self-energy in nuclear matter is  due to $S$- and

425: $P$-interaction of the pions and nucleons. The total pion self-energy

426: is thus a sum of a scalar and vector terms:

427: \begin{equation}\label{1f}

428: \Pi(\omega,k)\ =\ \Pi_S(\omega,k)+\Pi_P(\omega,k)\ .

429: \end{equation}

430: The main results of this paper are obtained for

431: $\Pi(\omega,k)=\Pi_P(\omega,k)$

432: operator. One can consider some models for $\Pi_S(\omega,k)$ and they

433: give the expressions like $\Pi_S = const\cdot\rho$ without $\omega$ and

434: $k$ dependence \cite{DRS,EW}. The influence  of $\Pi_S$ on

435: the final results has the quantitative character and is discussed in

436: Appendix A.

437: 

438: 

439: \subsection{$P$-wave part of the pion self-energy, $\Pi_P(\omega,k)$}

440: 

441: Following the papers \cite{Mi,PN,FW} we write $\Pi_P(\omega,k)$

442: as a sum of the

443: nucleon and isobar parts

444: \begin{equation}\label{2f}

445: \Pi_P(\omega,k)\ =\ \Pi_N(\omega,k)+\Pi_\Delta(\omega,k)\ .

446: \end{equation}

447: The operators $\Pi_N(\omega,k)$

448: and $\Pi_\Delta(\omega,k)$ can be expressed through

449: the lowest order polarization insertions

450: $\Pi^0_N(\omega,k)$ and $\Pi^0_\Delta(\omega,k)$

451: which can be presented as the integrals over the particle-hole ($\Phi_N$) and

452: isobar-hole loops ($\Phi_\Delta$). The vertices of these loops

453: are described by the pion-nucleon and isobar-nucleon-pion coupling constants

454: $f_{\pi NN}$, $f_{\pi N\Delta}$ and by the form

455:  factors  which take the nonzero baryon size into account,

456: $d_B(k^2)=\frac{\Lambda_B^2-m_\pi^2}{\Lambda_B^2+k^2}$,

457: where $B=N, \Lambda$ and $\Lambda_N$=0.667~GeV,

458: $\Lambda_\Delta$=1.0~GeV.  Polarization operators

459: $\Pi^0_N(\omega,k)$,

460: $\Pi^0_\Delta(\omega,k)$ can be written as:

461: \begin{equation}\label{3f}

462: \Pi^0_N(\omega,k) =

463: -4(\frac{f_{\pi NN}}{m_\pi})^2 k^2 d^2_N(k^2)\Phi_N(\omega,k),

464: \end{equation}

465: \begin{equation}\label{4f}

466: \Pi^0_\Delta(\omega,k)

467: = -\frac{16}9(\frac{f_{\pi N\Delta}}{m_\pi})^2 k^2

468: d^2_\Delta(k^2)\Phi_\Delta(\omega,k).

469: \end{equation}

470: Then for $\Pi_N$, $\Pi_\Delta$ we have the expressions \cite{DM}:

471: $$

472: \Pi_N(\omega,k) = \Pi^0_N(\omega,k)

473: \frac{1+C_0(\alpha(k)G'_{N\Delta} -G'_{\Delta\Delta})

474: \frac{-16}9\Phi_\Delta(\omega,k)}{E(\omega,k)} ,

475: $$

476: \begin{equation}\label{5f}

477: \Pi_\Delta(\omega,k)

478: = \Pi^0_\Delta(\omega,k)\frac{1+C_0(\frac1{\alpha(k)}G'_{N\Delta} -

479: G'_{NN})(-4)\Phi_N(\omega,k)}{E(\omega,k)}.

480: \end{equation}

481: In this equation $\alpha(k)=(\frac43f_{\pi N\Delta}

482: d_\Delta(k^2))/(2f_{\pi NN}d_N(k^2))$. In Eqs.~(\ref{5f}) we see

483: that $\Pi_N(\omega,k)$

484: is $\Pi^0_N(\omega,k)$ with one vertex renormalized due to

485: nucleon-hole and isobar-hole interaction.  The denominator $E(\omega,k)$

486: has a form

487: \begin{eqnarray}

488: E(\omega,k)&=&1+C_0\left(4G'_{NN}\Phi_N(\omega,k)

489: + \frac {16}9G'_{\Delta\Delta}\Phi_\Delta(\omega,k)\right.

490: \nonumber\\

491: &&\left. \quad +4\frac{16}9C_0(G'_{NN}G'_{\Delta\Delta} -

492:  G'^2_{N\Delta})\Phi_N(\omega,k)\Phi_\Delta(\omega,k)\right)

493: \label{6f}

494: \end{eqnarray}

495: and the condition

496: \begin{equation}\label{7}

497: E(\omega,k)=0

498: \end{equation}

499: gives a zero-sound dispersion equation. Solutions of this equation

500: describe the excitations in nuclear matter, if we do not take into

501: account the renormalization  due to the pion-baryon interaction.

502: Equation (1) has the solutions corresponding to the solutions of

503: Eq.~(\ref{7}). Thus at $G'_{NN}<0$, Eq.~(\ref{7}) has the

504: solutions $\omega_{sd}(k)$ (which will be shown and discussed

505: below), while at $G'_{NN}>0$  there are zero-sound solutions

506: $\omega_s(k)$ \cite{CC,PR,LaLi}. These solutions are the poles

507: of $\Pi_N(\omega,k)$ and $\Pi_\Delta(\omega,k)$ (\ref{5f}). The

508: equation (1) has the corresponding solutions:

509: $\omega^\pi_{sd}(k)$ and $\omega^\pi_s(k)$.  It can be seen from

510: the structure of Eq.~(1) that when  $f_{\pi NN}$ and $f_{\pi

511: N\Delta}$ approach zero points, the branches

512: $\omega^\pi_{sd}(k)$ ($\omega^\pi_s(k)$) approach to the poles

513: of polarization operator $\Pi(\omega,k)$, i.e., $\omega_s(k)$

514: ($\omega_{sd}(k)$).

515: 

516: \subsection{Extension of $\Pi^0_N(\omega,k)$ in $\omega$

517: to the unphysical sheets of the complex plane}

518: 

519: The causal polarization operator $\Pi^0_N(\omega,k)$

520: can be presented as a sum of two terms,

521:  the first  term corresponds to  excitation and the second to the

522: annihilation of the particle-hole pairs.

523: We want to study the excitations in nuclear matter

524:  which appear as a response on an external perturbation and we  use

525: the retarded operators in Eq.~(1). For the real $\omega$ they are

526: determined as \cite{LaLi,FW}

527: $$

528: \Pi^{0R}(\omega,k) = (Re + i\

529: sign(\omega) Im) \Pi^0(\omega,k)

530: $$

531: and $\Phi^R_N(\omega,k)$ is expressed through the integrals

532: \cite{FW}

533: \begin{equation}\label{9a}

534: \Phi^R_N(\omega,k) =

535: \frac{-1}{(2\pi)^3}\int d^3p\left[ \frac{\theta(|\vec p+\vec

536: k|-p_F)\theta(p_F-p)} {\omega+\frac{p^2}{2m}-\frac{(\vec p+\vec

537: k)^2}{2m}+i\eta} - \frac{\theta(p-p_F) \theta(p_F-|\vec p+\vec

538: k|)} {\omega+\frac{p^2}{2m}-\frac{(\vec p+\vec k)^2}{2m}+i\eta}

539: \right] .

540: \end{equation}

541: The function $\Phi^R_N(\omega,k)$ can be written as

542: \begin{equation}\label{9}

543: \Phi_N(\omega,k) =

544: \phi_N(\omega,k) + \phi_N(-\omega,k).

545: \end{equation}

546: The first integral in (\ref{9a}) corresponds to $\phi(\omega,k)$ and

547: the second one to $\phi(-\omega,k)$.

548: Equation (\ref{9a}) can be used to obtain  a condition

549:    which relates the values of $\Phi^R_N(\omega,k)$

550: in the two points of the complex plane \cite{LaLi}

551: \begin{equation}\label{1g}

552: \Phi^R_N(-\omega^*,k) = \Phi^{R*}_N(\omega,k).

553: \end{equation}

554: The same equation can be written for $\Phi^R_\Delta$ and, consequently,

555: for operators in Eqs.~(\ref{2f})-(\ref{6f}).  Henceforth only the

556: retarded operators are used and  index $R$ is omitted.

557: 

558: We write the expression for $\Phi_N(\omega,k)$  in such a way that it

559: does not contain the overlapping logarithm cuts (compare

560: with \cite{PN,FW}):

561: \begin{equation}\label{10}

562: \phi_N(\omega,k)=

563: \frac{m}{k} \frac1{4\pi^2}

564: \left(

565: \frac{-\omega m+kp_F}2-\omega m \ln\left(\frac{\omega m}{\omega

566: m-kp_F+\frac12k^2}\right)\right.

567: \end{equation}

568: $$

569: +\left.

570: \frac{(kp_F)^2-(\omega m-\frac12k^2)^2}{2k^2} \ln\left( \frac{\omega

571: m-kp_F-\frac12k^2}{\omega m-kp_F+\frac12k^2}\right)\right)

572: $$

573: for $0\le k\le2p_F$. Each of logarithms has its own region of

574: integration over $p$ and over the angle between $\vec p$ and $\vec k$

575: in (\ref{9a}).  These regions are shown in \cite{FW}.

576: 

577: At  $k\ge2p_F$,  $\phi_N(\omega,k)$ is the Migdal's function

578: \cite{Mi,EW}

579: \begin{equation} \label{11}

580: \phi_N(\omega,k) =\ \frac1{4\pi^2}\

581: \frac{m^3}{k^3}

582: \left[\frac{a^2-b^2}2 \ln\left(\frac{a+b}{a-b}\right)-ab\right]\,

583: \end{equation}

584: where $a=\omega-(k^2/2m)$, $b=kp_F/m$.

585: 

586: Function $\Phi_\Delta(\omega,k) = \phi_\Delta(\omega,k) +

587: \phi_\Delta(-\omega,k)$, where $\phi_\Delta(\omega,k)$ is determined by

588:  (\ref{11}) with $a=\omega-(k^2/2m) + (m_\Delta-m)$.

589: 

590: Let us consider the cuts of $\Pi^0_N(\omega,k)$ in the complex plane

591: of $\omega$. It is clear that, for $k\le2p_F$, there are two cuts (we

592: denote them by $I$ and $II$). The cuts originated by the

593: logarithmic terms of (\ref{10}) are at

594: \begin{equation} \label{12}

595: I: 0\ \le\ \omega\ \le\

596: \frac{kp_F}{m} -\frac{k^2}{2m}\,

597: \end{equation}

598: $$

599:  II:

600: \frac{kp_F}{m}-\frac{k^2}{2m}\ \le\ \omega\ \le\

601: \frac{kp_F}{m}+\frac{k^2}{2m}\ .

602: $$

603: The cuts of the function $\phi_N(-\omega,k)$ lie on the negative

604: semi-axis symmetrically with respect to the cuts in (\ref{12}).

605: Therefore, $\Pi^0_N(\omega,k)$ has four cuts in the complex

606: plane of $\omega$, which are shown in Fig.~2. When $\omega>0$,

607: the cuts are caused by the singularities of $\phi_N(\omega,k)$,

608: while at $\omega<0$ the cuts are related to the logarithms of

609: $\phi_N(-\omega,k)$.

610: 

611: The imaginary factors in denominators in  (\ref{9a}) determine the

612: physical sheet in the complex plane of frequency. On the physical sheet

613: the first logarithm in (\ref{10}) has the imaginary

614: part on the upper edge of the cut $I$ equal to $-\pi i$, and  $+\pi i$

615: on the lower edge of the cut.

616: In the point $\omega m-kp_F +\frac12k^2$=0, where the cuts

617: $I$ and $II$ touch each other, the imaginary part of

618: $\phi_N(\omega,k)$ is continuous but there is a jump of the

619: derivative. This is shown in \cite{FW}, Fig.~12.9.

620: 

621: Let us now determine the unphysical sheets which we pass on when go

622: under the cuts $I$ and $I'$ (Fig.~2). Consider the cut $I$.

623: We denote the first logarithm in (\ref{10}) as

624: $ln(z_1)\equiv ln\left(\frac{\omega m}

625: {\omega m-kp_F +\frac12k^2}\right)$.

626: While the frequency $\omega$ goes along the cut $I$ (\ref{12}):

627: $\omega=(0,\frac{kp_F}{m} -\frac{k^2}{2m})$ then $z_1$ is

628: changed in the interval $z_1=(0,-\infty)$. The cut $I$ in the

629: complex plane of $\omega$ corresponds to the  cut along the

630: negative real axis in the complex plane of $z_1$. When

631: we go to the cut $I$ from above (the arrow in Fig.~2a),  this

632: corresponds that we go to the cut from below in the complex plane

633: of $z_1$. Going over under the cut,

634:  we pass on to the unphysical sheet neighboring  with

635: the physical one on the Riemann surface of $ln(z_1)$. The

636: values of $ln(z_1)$ on the neighboring sheet differ by

637: of $(-2\pi i)$ from the values on the physical one at the same

638: $\omega$.

639: As example, the value of

640: $ln\left(\frac{\omega m}{\omega m-kp_F +\frac12k^2}\right)$ on the

641: lower edge of cut $I$ is

642: $ln\left(-\frac{\omega m}{\omega m-kp_F+\frac12k^2}\right) +\pi i$.

643: But the value at the same point  of

644: $\omega$ on the unphysical sheet is $ln\left(-\frac{\omega

645: m}{\omega m-kp_F +\frac12k^2}\right)-\pi i$.  Thus, we have the

646: continuous changing of logarithm along the arrow in Fig.~2a. Then

647: we conclude that $\Pi(\omega,k)$ in  Eq.~(\ref{1}) changes

648: continuously as well.

649: Thus, going over under the cut $I$ in Fig.~2a we pass on to the

650: unphysical sheet of the complex plane of $\omega$  that is

651: placed under the physical sheet. Let us denote the unphysical

652: sheet with the same letter as a cut: $I$.

653: The part of the unphysical sheet $I$ is

654: shown in Fig.~2b by the shading with the right  slope.

655: 

656: Now we define an unphysical sheet related to the cut $I'$ (denote

657: the sheet by $I'$). The cut $I'$ stems from the first logarithm

658: in $\phi(-\omega,k)$. We designate

659: $ln(z_1')\equiv

660: -ln\left(\frac{-\omega m}

661: {-\omega m-kp_F +\frac12k^2}\right)$ = $

662: ln\left(\frac{\omega

663: m+kp_F -\frac12k^2} {\omega m}\right)$.

664: Going over under the cut of $ln(z_1')$ we add the shift $-2\pi

665: i$ to the value of $ln(z_1')$. Then, the  logarithm

666: on the sheet $I'$  under the lower edge of the cut $I'$ is equal to

667: $ln\left(-\frac{\omega

668: m+kp_F -\frac12k^2} {\omega m}\right)+\pi i-2\pi i$.  This

669: expression is used for the analytical continuation to the sheet

670: $I'$. The part of the sheet $I'$ is shown on the Fig.~2b by the

671: shading with the left slope.

672: 

673: In the paper we consider the solutions of (\ref{1}) which are placed on

674: the shaded  parts of the unphysical sheets $I$ and $I'$ in

675: Fig.~2b. Each logarithmic term in (\ref{9}), (\ref{10}) has its

676: own Riemann surface.  An important point here is that moving on

677: Riemann surface of one of logarithms we stand on the

678: physical sheet for the other logarithms.

679: 

680: Thus, we have built up two unphysical sheets $I$ and $I'$. Let

681: us show that the values of $\Pi^0_N(\omega,k)$ are real and coincide

682: on the negative imaginary axes of the sheets $I$ and $I'$.

683: 

684:  On the imaginary axis we denote  $\omega=i\omega_i$.

685: On the negative imaginary axis of the sheet $I$, the function

686:  $\Phi_N$ (\ref{9}), (\ref{10}) takes a form

687: \begin{equation}\label{71}

688: \Phi_N(i\omega_i,k) = \frac {mp_F}{4\pi^2}\left[1 -

689: \frac{i\omega_i m}{kp_F} \left(\ln\frac{-i\omega_i m}{i\omega_i

690: m-kp_F +\frac12 k^2} - \pi i\right) \right.

691: \nonumber

692: \end{equation}

693: $$ \left.

694: -\frac{i\omega_i m}{kp_F} \left(\ln\frac{i\omega_i m+kp_F

695: -\frac12 k^2}{i\omega_i m}\right) + \ ... \right].

696: $$

697: \begin{equation}\label{81}

698: = \frac {mp_F}{4\pi^2}

699: \left[1 +2\frac{\omega_i m}{kp_F}

700:     \left(arctg\frac{\omega_im}{kp_F-\frac12 k^2} - \frac\pi 2\right)

701: +  ...\right] \ .

702: \end{equation}

703: Here only the first and the second terms of $\phi(\omega,k)$ and

704: $\phi(-\omega,k)$ (\ref{10}) are shown.

705: 

706: On the other hand, on the  negative imaginary axis of the sheets $I'$

707: the function $\Phi_N$ (\ref{9}) is

708: \begin{equation}\label{81a}

709: \Phi_N(i\omega_i,k) = \frac{mp_F}{4\pi^2}

710: \left[1 -\frac{i\omega_i m}{kp_F}

711: \ln\frac{i\omega_i m}{i\omega_i m-kp_F +\frac12 k^2}

712: \right.

713: \nonumber

714: \end{equation}

715: $$ \left.

716: -\frac{i\omega_i m}{kp_F}

717: \left(\ln\frac{i\omega_i m+kp_F -\frac12 k^2}{-i\omega_i m}

718:     - \pi i\right) +  ... \right]

719: $$

720: and after reduction we have the expression (\ref{81}) again.

721: This is the illustration of the following connection between  the

722: values of $\Pi^0_N(\omega,k)$ on  $I$ and $I'$  sheets:

723: \begin{equation}\label{6a}

724: \left(\Pi^{0}_N(\omega,k)\right)^*_I =

725: \left(\Pi^{0}_N(-\omega^*,k)\right)_{I'} .

726: \end{equation}

727: 

728: Thus, $\Pi^0_N(\omega,k)$  is defined on the lower semiplanes of the

729: unphysical sheets $I$ and $I'$ (the shaded regions in Fig.~2b),

730: and the values of $\Pi^0_N(\omega,k)$ coincide on the negative imaginary

731: axes of the sheets $I$ and $I'$.  The special solutions of (1)

732: which we are interested in, are placed on the shaded part of the

733: complex plane of $\omega$ in Fig.~2b and on the imaginary axes of

734: the physical and unphysical sheets.

735: 

736: \section{Investigation of instability at $G'_{NN} < G'_{tr}$}

737: 

738: In this part we consider solutions of the pion dispersion equation (1)

739: at $G'_{NN} < G'_{tr}$. The definition of $G'_{tr}$ will be given in

740: the Sect.~5, since it is determined by the behavior of

741: branches  which are studied in

742: Sects.~3,4. We start by considering the zero-sound dispersion

743: equation, and investigate its  solutions for different values of

744: $G'_{NN}$. We demonstrate the modification of these solutions when

745: we pass to the pion dispersion equation.

746: 

747: 

748: 

749: \subsection{Zero-sound dispersion equation}

750: 

751: In this Subsection we consider the unstable

752: solutions $\omega_{sd}(k)$ of the

753: zero-sound dispersion equation. For $G'_{NN}<-1$,

754: they are amplified, placed on the physical sheet and

755: reflect the instability of nuclear matter in respect to the

756: small fluctuations of density. For $-1<G'<0$, the branches

757: $\omega_{sd}(k)$ are damped, and come to  the  negative

758: imaginary axes (which belongs to the unphysical sheets $I$ and

759: $I'$).

760: 

761: In this Section we do not take into account the excitations

762: of isobar and restrict ourselves by

763: $\Pi(\omega,k)=\Pi_P(\omega,k)$ in (\ref{2f}). The

764: zero-sound dispersion equation stems from (\ref{7}) and has the

765: form

766: \begin{equation}\label{17x}

767:  -4C_0\Phi_N(\omega,k)=\frac1{G'} .

768: \end{equation}

769: The results obtained for this equation are valid both in

770: spin-isospin and in isoscalar, spin, isoscalar channels.

771: 

772: First we obtain the solutions of Eq.~(\ref{17x}) in the

773: kinetic theory of Landau  and compare them with the known

774: results. The function $\Phi_N(\omega,k)$ (\ref{9},\ref{10}) in

775: the long wavelenths limit is

776: \begin{equation}\label{13a}

777: \Phi_N(\omega,k) = 2 \frac{m p_F}{4\pi^2}\left(1-

778: \frac{\omega m}{2 k p_F}

779: \ln \frac{\frac{\omega

780: m}{ k p_F} +1}{\frac{\omega m}{ k p_F}-1}

781: \right) .

782: \end{equation}

783: Substituting this expression in (\ref{17x}), we

784: obtain the known zero-sound dispersion equation  \cite{LaLi}

785: \begin{equation}\label{5x}

786: 1+\frac 1{G'} = \frac{\omega m}{2kp_F}ln\left(\frac

787: {\omega m + kp_F}{\omega m - kp_F}\right) =

788: \frac{s}{2}\ln\frac{s+1}{s-1} .

789: \end{equation}

790: Here we denote $s=\frac{\omega m}{kp_F}$.

791: When $G'<-1$, this equation has two (positive and negative) solutions

792: on the imaginary axis of the physical sheet

793: \cite{LaLi,PN}.  Let us denote $s=i\gamma$, then

794: Eq.~(\ref{5x}) transforms into

795: \begin{equation}\label{6x}

796: 1+\frac 1{G'} = \gamma\ arctg\frac 1{\gamma}\ .

797: \end{equation}

798: 

799: To obtain solutions for $-1<G'<0$, we take   the long

800: wavelength limit of Eq.~(\ref{81}) and substitute it in (\ref{17x}). On

801: the negative imaginary axis of the unphysical sheet,

802:  we get an equation

803: \begin{equation}\label{9x}

804: 1 + \frac 1{G'} = -\gamma(arctg(\gamma) - \frac {\pi}{2}) .

805: \end{equation}

806:  Solutions of this equation are negative, i.e., $\gamma<0$.

807: This equation coincides with that presented in paper \cite{CC}.

808: It was pointed out in \cite{CC} that the solution of

809: Eq.~(\ref{9x}) `` do not correspond to an actual solution". In

810: agreement with this statement we see that solutions of (\ref{9x})

811: are located on the negative imaginary  axis belonging both $I$ and

812: $I'$ unphysical sheets.  It was explained in \cite{CC} the damping of

813: this mode is the Landau damping \cite{LaLi}.

814: 

815: 

816: 

817: In Fig.~3 the solutions $\gamma$ to Eq.~(\ref{17x}) are shown at

818: $G'<0$. They are similar to the solutions at $G'<0$ in Fig.~1 of

819:  \cite{CC} (except the curve $ab$  at $-1<G'<0$). The shaded

820: region means that the solution are on the unphysical sheets. In

821: Fig.~3 it is demonstrated that when the attraction between the

822: particles is increasing, the damping solution placed on the

823: unphysical sheet goes over to the physical sheet, turning to

824: amplified solution.

825: 

826: Considering  the unphysical sheets we can find two solutions of

827: (\ref{17x}) for any value of $G'<0$ (Fig.~3). However, at

828: $-1<G'<0$ they are on the different unphysical sheets.  It was

829: shown that the solutions of Eq.~(\ref{9x}) are negative thus

830: being located on the unphysical sheet $I$.  Now we  look for

831: solutions with $\gamma>0$ (curve $ab$ in Fig.~3). The situation

832: is the same as we had with the sheets $I$ and $I'$. There are

833: two new unphysical sheets $\tilde I$ and $\tilde I'$ which

834: overlay the physical one. In Appendix B  the construction of

835: these sheets is presented. The dispersion equation (\ref{17x})

836: on the imaginary positive unphysical axis is the same on the

837: both $\tilde I$ and $\tilde I'$ sheets and has a form

838: 

839: \begin{equation}\label{13}

840: 1 + \frac 1{G'} = -\gamma(arctg(\gamma) + \frac {\pi}{2}) .

841: \end{equation}

842: The solutions of this equation are positive. This equation turns

843: into Eq.~(\ref{9x}) if to change the sign of $\gamma$. The solutions of

844: this equation are shown by the curve $ab$ in Fig.~3. They are located on

845: the positive imaginary unphysical  axis

846:  (this is marked by the  horizontal shading).

847: 

848: Let us compare Fig.~3 and Fig.~1 in \cite{PR}. In paper \cite{PR} only

849: one curve is presented which corresponds to decreasing of $\gamma$ with

850: increasing of $G'$. This curve describes  the physical processes: the

851: amplifying of the amplitudes of the small fluctuations at $G'<-1$ and

852: the Landau damping of excitations at $-1<G'<0$. According to

853: \cite{Mi,BaW} in the spin-isospin channel this curve can be considered

854: as describing the $\pi^+$-type excitations. Another curve shown in

855: Fig.~3 corresponds to the $\pi^-$-type excitations. Their frequencies

856: are to be taken with the opposite sign in respect to the ones shown in

857: Fig.~3.  There are two solutions  for $\pi^0$-type excitations.  One of

858: them corresponds to the growing excitations (at $G'<-1$) and the second

859: should be taken with the opposite sign and becomes to degenerate with

860: the first one in the symmetric and asymmetric matter.

861: 

862: It is well known that the kinetic theory of Landau gives

863: the linear dependence of

864: frequencies of  collective excitations  on the  wave

865: vectors, i.e., $\omega\sim k$. When we pass

866: to RPA using Eqs.~(\ref{9}), (\ref{10}), the shape of

867: the dependence changes and we obtain the branches of solutions

868: $\omega_i(k)$.  In Fig.~4 we show the imaginary branches

869: $\omega_{sd}(k)$ obtained in RPA.  The shaded fields on the

870: edges of figures indicate that the branches are located on the

871: unphysical imaginary axis. We see that the branches change

872: continuously with $G'$.  When  $-1 < G' < 0$, the branches

873: $\omega_{sd}(k)$ are located on the unphysical sheet only, but for $G'

874: < -1$  they go over to the physical sheet in the interval of wave

875: vectors $k=(0,k_{fin})$.

876: 

877: In analogy with Fig.~3, we can find the second solution of

878: Eq.~(\ref{1}) for each values of $G'$ and $k$. The ``symmetric"

879: branch has the conjugated values on the physical sheet

880: $\omega_{sd}^*(k)$, but it goes to the unphysical sheet $\tilde

881: I$ when $k>k_{fin}$.

882: 

883: All solutions belong to the same family $\omega_{sd}$. This

884: family exists at $G'<0$.

885: 

886: \subsection{Pion dispersion equation}

887: We consider the influence of the pion-nucleon

888: interaction on the branches $\omega_{sd}(k)$. Following \cite{Mi}

889: we  include only the direct term of the interaction of the pion with a

890: particle-hole pair, and do not take into account the noncentral

891: interaction of the quasiparticles. In this Section we take into account

892: the isobar-hole pairs contribution to the pion self-energy,

893: Eq.~(\ref{7}).

894: 

895: First let us investigate how the contribution of isobar-hole

896: excitations changes the stability conditions. Let us assume

897: in Eq.~(\ref{17x}) that $\omega$=0 and $k\to 0$. In this limit

898: this equation reduces to $-1=\frac1{G'_{NN}}$.  This gives the

899: value of $G'_{NN}$ when instability occurs in the long

900: wavelength limit.  If we take into account the isobar-hole

901: loops, this condition takes the form

902:  \begin{equation}\label{17xy}

903:  -4C_0\Phi_N(\omega=0,k)\approx

904: \frac{1+ C_0\frac{16}9\Phi_\Delta(0,k) G^{'2}_{\Delta\Delta}}

905: {G'_{NN}(1+C_0\frac{16}9\Phi_\Delta(0,k) G'_{\Delta\Delta})

906: - C_0\frac{16}9\Phi_\Delta(0,k) G^{'2}_{N\Delta}}.

907: \end{equation}

908: 

909: At $\omega$=0 and $k\rightarrow 0$, $\Phi_\Delta(0,k)$ is

910:  proportional to nuclear density: $\Phi_\Delta(0,k)\approx

911: \frac{\rho}{2(m_\Delta-m)}$. If $\rho \rightarrow 0$, the

912:  corrections produced by $\Phi_{\Delta}$ disappear. As a result

913: the values of  $\rho_c$ and $k_c$ on the solid lines in Fig.~1

914: approach the dashed ones. In Fig.~1 the solid lines  are obtained

915: for the minimal densities when solutions with $\omega$=0 emerge

916: for every $G'_{NN}$.

917: 

918: It is shown in Fig.~5a, how   $\omega^\pi_{sd}(k)$

919: are modified when $f_{\pi NN}$ is changed  from 0 to 1. At

920: $f_{\pi NN}$=0 the branch of solutions of the zero-sound

921: dispersion equation, Eq.~(\ref{7}), $\omega_{sd}(k)$  is

922: presented. (It should be noted that the results in  Fig.~5a are

923: obtained for $\Pi_\Delta(\omega,k)$ contribution being  included

924: in pion self-energy (\ref{2f}) while those in Fig.~4 do not

925: include $\Pi_\Delta(\omega,k)$.) As we see from Fig.~5a, the

926: growth of $f_{\pi NN}$ results in emerging of the amplified

927: solutions on the physical sheet at $-1< G'_{NN}< G'_{tr}$.

928: 

929: In Fig.~5b the branches $\omega^\pi_{sd}(k)$ are shown for different

930: values of density of nuclear matter. The curve 3 is calculated for

931: $\rho=\rho_c$ and  contains the point

932: $\omega^\pi_{sd}(k_{c})$=0.  It is demonstrated how the zero

933: value $\omega$=0 is placed on this branch (the point $A$).  At

934: the point $A$ the condition

935: $\left(\frac{d\omega^\pi_{sd}(k)}{dk}\right)_{k=k_c}$=0 is

936: satisfied.  The density $\rho_{c}$ and wave vector $k_{c}$,

937: corresponding to  zero solution, are presented in Fig.~1 (denoted by

938: $A$).  When density is lower than the critical one, $\omega^\pi_{sd}(k)$

939: are completely situated on the unphysical sheet (curve 4) and the

940: nuclear matter is stable. When $\rho>\rho_c$, the amplified solutions

941: appear on the physical sheet in the certain interval of $k$ (the curves 1

942: and 2).

943: 

944: Figures~5a,~b demonstrate the typical behavior of the branches

945: of solutions of (1) at $G'_{NN} <G'_{tr}$. Thus, we obtain the

946: instability of nuclear matter related to $\omega^\pi_{sd}(k)$

947: for all $G'_{NN} <G'_{tr}$. Instability occurs at

948: $\rho\geq\rho_c$.  The corresponded solutions of the zero-sound

949: dispersion equation, $\omega_{sd}(k)$, demonstrate instability

950: at $G'_{NN} < -1$ only.

951: 

952: We refer $\omega^\pi_{sd}(k)$ to $\omega_{sd}$ family.

953: 

954: 

955: \section{Investigation of instability at $G'_{NN} > G'_{tr}$}

956: 

957: Now let us consider in Fig.~1 another well studied region

958: $G'_{NN}\approx 2$. This region of $G'_{NN}$ was investigated in

959: connection with prediction of the pion condensation.

960: 

961: In Fig.~6 the solutions of Eq.~(1) are presented  at $G'_{NN}=2$

962:  and $\frac{\rho}{\rho_0}$=1.78. Such figures were published in

963: papers \cite{Mi}. In Fig.~6 we see the pion branch of solutions

964: $\omega_\pi(k)$, it starts at the frequency

965: $\frac{\omega_\pi(k=0)}{m_\pi}=1$. It turns into the free pion

966: solution as $f_{\pi NN}\rightarrow 0$.  The isobar branch

967: $\omega_\Delta(k)$ starts at the frequency which is determined

968: by the isobar and nucleon mass difference. The zero-sound branch

969: $\omega^\pi_s(k)$ starts at $\omega_s(k=0)$=0, and at certain

970: value of $k$ it acquires an imaginary part due to  Landau

971: damping \cite{LaLi}. It goes over  to unphysical sheet related

972: to the cut $II$, Eq.~(\ref{12}), \cite{DRS,SR}.  Besides the

973: numerated branches, one more imaginary branch is presented in

974: Fig.~6. Let us denote it $\omega_c(k)$.  Tracing of this branch

975: of solutions stimulated  development of the theory of the ``pion

976: condensation" \cite{Mi}.

977: 

978: In this Section we investigate the family of solutions

979: $\omega_c$ and show that this family is responsible for the

980: instability at $G'_{NN} > G'_{tr}$. In Fig.~6  a part of a

981: branch $\omega_c(k)$ is presented that is located on the

982: physical sheet in the interval  $k_1 \leq k\leq k_2$. The rest

983: of this branch is on the unphysical sheet $I$.

984: 

985: In Fig.~7 we present the  same branch $\omega_c(k)$  as in

986: Fig.~6, but here we show it from another point of view,

987: i.e., considering the complex plane of $\omega$ (the solid line).

988: The branch $\omega_c(k)$ starts at $k$=0 at the point `1' and

989: has the real value $\frac{\omega_c(k=0)}{m_\pi} = 1$. With the

990: increasing of $k$ it moves to the unphysical sheet $I$. It

991: follows from (\ref{6a}) that there is another branch of

992: solutions $\omega'_c(k)$, which starts at the point

993: $\frac{\omega'_c(k=0)}{m_\pi}=-1$ and is placed on the

994: unphysical sheet $I'$ (the dashed curve).

995: 

996: Below in paper, the hatching on the edge of figures with the

997: right slope means that the drawing branches are located on the

998: unphysical sheet $I$. The left slope means that they are on the

999: sheet $I'$. The physical sheet is above the axis of abscissas.

1000: 

1001: In point `2' $\omega_c(k)$ and $\omega'_c(k)$ coalesce on

1002: the imaginary axis. Then $\omega_c(k)$ ascends  the imaginary

1003: axis and $\omega'_c(k)$ descends.

1004: (To determine which of branches go up we can

1005:   take into account the width of isobar in Eq.~(\ref{4f}) as in

1006: \cite{DRS}.) Thus, two simple roots of Eq.~(1) which move on the

1007: unphysical sheets $I$ and $I'$ (when $k$  increase) coalesce on the

1008: imaginary axis at $k=k_m$. Here the roots become imaginary ones,

1009: and continue to move up and down the imaginary axis which belongs

1010: to  both $I$ and $I'$  sheets.

1011: 

1012: In Fig.~7 the values of $\omega_c(k)$ and $\omega'_c(k)$ are shifted

1013: from the imaginary axis by hand to demonstrate the trajectories of

1014: solutions with increasing of $k$ (it is shown by arrows). The wave

1015: vector $k=k_t$  corresponds to the turning point of

1016: the trajectories `3'. At this point, $\omega_c(k)$ and $\omega'_c(k)$

1017: reach their  maximal and  minimal values. When $k >k_t$, the

1018: branches return along the imaginary axis.  In Fig.~7 the

1019: solutions are presented for $k\leq 2.45 m_\pi$. At larger $k$,

1020: the behavior of branches has a nontrivial character but it is

1021: not related to physical processes and is not discussed here. In

1022: Figs.~6,~7 it is shown how 1)~the branch $\omega_c(k)$ goes

1023: from the unphysical  to the physical sheet at $k=k_1$; 2)~it

1024: becomes imaginary, with growing value of $Im(\omega_c(k)) > 0$,

1025: on the physical sheet; 3)~the branch returns to the unphysical

1026: sheet at $k=k_2$.

1027: 

1028: In Fig.~8  the complex plane of $\omega$ is presented where

1029: $\omega_c(k)$ and $\omega'_c(k)$ are shown for different densities of

1030: nuclear matter. In these figures (except Fig.~8a), the

1031: branches are drawn up to the turning point on the imaginary

1032: axis, i.e., at $k=(0,k_t)$ (see Fig.~7).  In Fig.~8a the density

1033: is small and the branches do not attain the imaginary axis. In

1034: Fig.~8b the branches encounter on the imaginary axis but

1035: completely stay on the unphysical sheets ($k_t=1.78m_\pi$). In

1036: Fig.~8c  the result for the critical density $\rho_c$  is

1037: presented. On the real axis the solution with $\omega$=0

1038: appears. Here $k_t=k_c$ ($k_t=1.81m_\pi$)  and we see that the

1039: conditions $\omega_c(k_c)$=0 and

1040: $\left(\frac{d\omega_c(k)}{dk}\right)_{k=k_c}$=0 are satisfied.

1041: The wave vector and density corresponding to the point $B$ in Fig.~8c

1042: are marked by $B$ in Fig.~1. In Fig.~8d it is demonstrated that at

1043: $\rho>\rho_c$ the solutions with $Im(\omega_c(k)) > 0$

1044: ($k_t=1.88m_\pi$) exist on the physical sheet.

1045: 

1046: The critical densities and wave vectors in Fig.~1 for $G'_{NN} >G'_{tr}$

1047: correspond to the solutions denoted by $B$ in Fig.~8c. Such solutions

1048: $\omega_c(k)$ can be  found not only when $G'_{NN} > G'_{tr}$,

1049: but for $G'_{NN} < G'_{tr}$ also. It should be mentioned, that an

1050: attempt to shift the pion condensation to  larger densities means that

1051: $\omega_c(k)$ go on  to stay on the unphysical sheet at

1052: larger densities. There are several reasons that can influence our

1053: results. It was shown in \cite{DRS} that the important factor shifting

1054: the pion condensation to the larger densities is the decreasing of the

1055: effective nucleon mass with growth of density. This decreasing is

1056: predicted by the scaling of Brown-Rho \cite{BR} and QCD sum rules

1057: \cite{MRL,DRSF}.  However, at smaller values of the nucleon mass

1058: the lowest order expansion in powers of $\frac{p^2}{2m}$ in

1059: Eq.~(\ref{9a}) is not valid any more, thus requiring relativistic

1060: kinematics for the nucleon.  Besides, the formation of isobar Fermi

1061:  surface must be taken into account, as soon as the density becomes so

1062: large that the quasiparticle energy on the Fermi surface exceeds the

1063: difference of the isobar and nucleon  mass.

1064: 

1065: In Figs.~7,~8 the branches $\omega_c(k)$ start at the point

1066: $\omega_c(k=0)=m_\pi$. Inclusion of the scalar part of the pion

1067: self-energy $\Pi_S$, Eq.~(\ref{1f}),  displaces the

1068: initial value to $\omega^2_c(k=0)=m^2_\pi + \Pi_S$ (Appendix A).

1069: However, this does not modify the curves in Fig.~1

1070: qualitatively.

1071: 

1072: 

1073: \section{Determination of $G'_{tr}$}

1074: 

1075: In Sects.~3, 4 we have presented  zero-frequency solutions

1076: of Eq.~(1), which appeared for different values of $G'$.

1077:  In this Section the ``transitional" value  $G'_{tr}$

1078: is obtained.  In Sects.~3,4 it was shown

1079: that at $G'_{NN} < G'_{tr}$ the zero-frequency solutions belong to

1080: the family $\omega_{sd}$ and  at $G'_{NN} > G'_{tr}$ they

1081: refer to  $\omega_c$. In this Section the

1082: transition of solution with $\omega$=0 from one family  to

1083: another is demonstrated.

1084: 

1085: In the vicinity of $G'_{tr}$,

1086: at $G'_{NN} \approx G'_{tr}$, the branches belonging to

1087:  $\omega^\pi_{sd}$ and $\omega_c$ families are placed

1088: close to each other on the same unphysical sheets. In Figs.~5,~7,~8

1089: these branches are shown separately. In Fig.~9 the branches of the both

1090: families are shown simultaneously.

1091: 

1092: 

1093: 

1094: In Fig.~9a at $G'_{NN}=-0.0020$ and $\rho=\rho_c$,

1095: the branches $\omega_c(k)$, $\omega'_c(k)$ and

1096: $\omega^\pi_{sd}(k)$ are presented simultaneously.

1097: The critical value of the wave vector is

1098: $k_c=1.61m_\pi$. The branches $\omega_c(k)$ and $\omega'_c(k)$ behave

1099: like  in Fig.~8a. The branch $\omega^\pi_{sd}(k)$ is shown

1100: by the dotted curve and it is presented in the interval  $k=(0, k_c)$.

1101: The behavior of $\omega^\pi_{sd}(k)$ is similar to the curve 3 in

1102: Fig.~5b between $k$=0 and the point $A$, which corresponds to

1103: $k=k_c$.

1104: 

1105: 

1106: In Fig.~9a number `1' denotes the start of branches at $k$=0;

1107: `2' means the turning point of

1108: $\omega^\pi_{sd}(k)$:~$k_t$=0.077$m_\pi$; `3' corresponds to

1109: $k$=0.1$m_\pi$. We see that $\omega^\pi_{sd}(k)$ descends when

1110: $k$ changes from '0' to $k_t$ and then ascends and turns to zero

1111: at $k=k_c$:~$\omega^\pi_{sd}(k=k_c)$=0 (point $C$ in Fig.~1).

1112: 

1113: Now let us enlarge a little the coupling

1114: constant:~$G'_{NN}$=-0.0018.  The following calculations are

1115: made at the critical density $\rho=\rho_c$, corresponding to

1116: this value of $G'_{NN}$. In this case we have a coalescence of

1117: $\omega_c(k)$ and  $\omega'_c(k)$ on the imaginary axis. The

1118: wave vector of coalescence is equal to $k_m$=0.0962$m_\pi$. In

1119: this difficult for drawing case, the presentation of the

1120: branches is made in two steps. In Fig.~9b the branches are shown

1121: at $k\leq 0.094m_\pi$, i.e., before coalescence and in Fig.~9c at

1122: $k\geq 0.094m_\pi$. The position of branches at $k$=0.094$m_\pi$

1123: are marked by `2'.

1124: 

1125: In Fig.~9b it is shown that the position of  $\omega^\pi_{sd}(k)$ at

1126: point `2' is much lower on the negative imaginary axis than the

1127: $Im\omega_c(k)$ and  $Im\omega'_c(k)$. Further, in Fig.~9c we see the

1128: coalescence of $\omega_c(k)$ and $\omega'_c(k)$ at $k=k_m$ (point `3').

1129: Then $\omega_c(k)$ and $\omega'_c(k)$ go in the opposite directions

1130: along the imaginary axis.

1131: 

1132: In Fig.~9c it is shown that the branch $\omega'_c(k)$ descends and

1133: meets the branch $\omega^\pi_{sd}(k)$ in the point `4' (at $\tilde

1134: k_m$=0.109$m_\pi$).

1135: [The branch $\omega^\pi_{sd}(k)$ has turned upward at

1136: $k$=0.104$m_\pi$.] After the second  coalescence, the branches

1137: $\omega'_c(k)$ and $\omega^\pi_{sd}(k)$ go in the opposite sides

1138: on the complex plane. Thus, the branch $\omega'_c(k)$ has

1139: blocked $\omega^\pi_{sd}(k)$ on the unphysical sheet.

1140: 

1141: The branch $\omega_c(k)$ goes over to the real axis and a solution with

1142:  $\omega$=0 (point $D$) belongs to this

1143: branch:~$\omega_c(k_c)$=0 at $k_c$=1.60$m_\pi$ (the point $D$ in

1144: Fig.~1).

1145: 

1146: Thus, the value of $G'_{tr}$  is determined by the interaction of

1147: branches on the unphysical sheets shown in Fig.~9. We can

1148: conclude that $G'_{tr}$ is a special value of the spin-isospin

1149: effective quasiparticle interaction  when 1)~there is  coalescence of

1150: $\omega_c(k)$ and $\omega'_c(k)$ on the imaginary axis at the critical

1151: density $\rho=\rho_c$  and 2)~$Im\omega^\pi_{sd}(k_m)\leq$

1152:  $Im\omega_c(k_m)$, $Im\omega'_c(k_m)$. The second condition is

1153: satisfied as a rule, since $Im\omega^\pi_{sd}(k_m)$ tends to $-\infty$

1154: as $G'\rightarrow -0$ (compare with Fig.~3).

1155: 

1156: In Fig.~9 the results  are presented for

1157: $\Pi(\omega,k)=\Pi_N(\omega,k)$ (\ref{1f}),

1158: i.e., without the excitation of isobars. In this case we have obtained

1159: $G'_{tr}$=-0.0019.  If to solve Eq.~(1) with

1160: $\Pi_P(\omega,k) = \Pi_N(\omega,k) + \Pi_\Delta(\omega,k)$,

1161: then $G'_{tr}$=0.0020. The

1162: influence of $S$-wave  pion-nucleon interaction is discussed in

1163: Appendix A.

1164: 

1165: It is interesting to mention that the figure similar to Fig.~9 is

1166: presented in \cite{PR} for the solutions of the zero-sound dispersion

1167: equation. Their figure demonstrates the

1168: behavior of solutions for the different values of $F_0$ and the

1169: branches of solutions depend on $F_1$ instead of $k$ (this means the

1170: dependence on the value of the quasiparticle effective mass $m^*$ along

1171: the branch).  In paper \cite{PR} $F_0$ and $F_1$ are the zero and the

1172: first multipolar components of the scalar-isoscalar effective

1173: interaction of the quasiparticles.

1174: 

1175: 

1176: \section{ Discussion}

1177: 

1178: \noindent

1179: {\bf A.} In this paper two families of solutions to the pion dispersion

1180: equation are calculated, which determine the  stability boundary

1181: of the nuclear matter  (the critical density $\rho_c$)

1182: depending on the value of spin-isospin coupling constant $G'$.

1183: It is shown that there is  a special value of $G'= G'_{tr}$

1184: such that for $G'< G'_{tr}$ the stability boundary are related

1185: to the family $\omega^\pi_{sd}$ and  for $G'> G'_{tr}$ it

1186: refers to $\omega_c$.

1187: 

1188: It is interesting to investigate, whether we can  obtain the

1189: information about the difference of the phase transitions  at

1190: $G'<G'_{tr}$ and $G'>G'_{tr}$ using the  branches of solutions.

1191: To do this let us consider the rate of growth of excitations in the

1192: vicinity of $G'_{tr}$. For $\rho > \rho_c$,  Eq.~(1) has

1193: solutions with $Im \omega_i(k) > 0$. In Figs.~5,~7,~8d such

1194: solutions are presented. We determine $\Gamma(k)=-i\omega_i(k)$,

1195: $i=sd,c$.  We can take several values of $G'$ close to

1196: $G'_{tr}$ and calculate the  maximal values of $\Gamma(k)$

1197: (i.e., $\Gamma(k_t)$) depending on $\rho$ for the every $G'$.

1198: In Fig.~10 these results are demonstrated for

1199: $G'=G'_{tr}, G'_{tr}\pm 0.02, G'_{tr}\pm 0.2$, where

1200: $G'_{tr}$=-0.0019.  In this figure we cannot observe the essential

1201: difference in  the dependence of $\Gamma(k_t)$ on $\rho$ for

1202: $G'<G'_{tr}$ and $G'>G'_{tr}$. It is possible to conclude that either

1203: this test is not sensitive to the type of excitations, or the character

1204: of the phase transition  changes continuously with $G'$ from long

1205: wavelength instability up to the pion condensation.

1206: 

1207: \noindent

1208: {\bf B.} Now let us consider the singularity of the integrals

1209: containing the pion propagator $D$ (1): $F=\int

1210: D(\omega,k)f(\omega,k)d\omega d^3k$, in the poles of $D(\omega,k)$ when

1211: $\rho=\rho_c$ and $k\to k_c$.  The function $f(\omega,k)$ is not

1212: related to the poles of $D(\omega,k)$. In the Sect.~3.1 for the

1213: zero-sound dispersion equation (\ref{7}), it was shown that

1214: there are two solutions on the physical sheet. At $-1<G'<0$, the

1215: amplified solutions go over to the unphysical sheets $I$, $I'$

1216: and the second branch goes over to the unphysical sheets $\tilde

1217: I$, $\tilde I'$ (Fig.~3).

1218: 

1219: Similar situation takes place for the pion dispersion equation. Up to now

1220: we considered the solutions which are located on the physical and

1221: unphysical $I$ and $I'$ sheets, but there are the ``symmetrical"

1222: solutions. They have the conjugated values on the physical sheet and

1223: their unphysical sheets are $\tilde I$ and $\tilde I'$. This means that

1224: at $\rho=\rho_c$ and  $k=k_c$ not only $\omega_i(k_c)$=0

1225: ($i=sd,c$) but the ``symmetrical" branch as well. As a result,

1226: the function $F$ has a form (we leave the singular part only) $$

1227: F\approx \int \frac{d\omega

1228:  d^3k}{(\omega^2-\omega^2_i(k))^2}f(\omega=0,k_c).

1229: $$

1230: Since we are interested in such zero solutions that

1231: $\left(\frac{d\omega_i(k)}{dk}\right)_{k=k_c}$=0, then

1232: $\omega_i=a(k-k_c)^2$ as  $k\to k_c$ and the integral diverges along

1233: the curves in Fig.~1. This correspond to situation of the

1234: phase transitions \cite{LaLi}.

1235: 

1236: \noindent

1237: {\bf C.} In the paper we do not consider the case when

1238:  $G'_{N\Delta}$, $G'_{\Delta\Delta}$ are modified

1239: simultaneously with $G'_{NN}$. But

1240: as it is shown above, the existence of $G'_{tr}$  and

1241: $\omega_c(k)$, $\omega_{sd}(k)$ does not depend on whether we take

1242: excitation of the isobar into account or not. Therefore we do

1243:  not expect the qualitative changes in our results if we change

1244: the values of $G'_{N\Delta}$ and $G'_{\Delta\Delta}$ together with $G'_{NN}$.

1245: 

1246: 

1247: \section{ Conclusion}

1248: 

1249: In the paper we investigate of the zero-frequency solutions of the

1250: pion dispersion equation depending on the values of the spin-isospin

1251: coupling constant of quasiparticles $G'$. For every

1252: $G'$,  the critical density $\rho_c$ and the wave vector

1253: $k_c$  corresponding to the solution with $\omega$=0 were

1254:  calculated.  Then the branches of solutions, containing these

1255: zero solutions, are constructed and it is determined to what

1256: type of excitations (to what family) these  solutions referred.

1257: 

1258: It is shown that the zero-frequency solutions belong to the

1259: different types of excitations, depending the value of $G'$. In

1260: all models used in calculations, the ``transitional" value

1261: $G'_{tr}$ ($|G'_{tr}|\ll 1$) is found, such that at $G'=G'_{tr}$

1262: the zero solution passes from the branches of the type

1263: $\omega_{sd}$ to the ones of type $\omega_c$.

1264: 

1265: At $G' < G'_{tr}$, the unstable amplified branches of solutions

1266: for the zero-sound dispersion equation ($\omega_{sd}(k)$) and

1267: for the pion dispersion equation ($\omega^\pi_{sd}(k)$) are

1268: constructed (Figs.~3,~4,~5).

1269: In Fig.~5a we see that if to take the

1270: pion-nucleon coupling constant $f_{\pi NN}$ value to be very small,

1271: $\omega_{sd}(k)$ and $\omega^\pi_{sd}(k)$ are very close. When we

1272: increase $f_{\pi NN}$ up to 1, the behavior of $\omega^\pi_{sd}(k)$ is

1273: change: the damping solutions become amplified ones (Fig.~5). We obtain

1274: that the zero-frequency solutions of Eq.~(1) at $G' < G'_{tr}$ belong

1275: to the same family of solutions which is responsible for the long

1276: wavelength instability in the kinetic theory of Landau.

1277: 

1278: Figure 1 demonstrates that the stability boundary continuously

1279: extends to $G' > G'_{tr}$. But at these $G'$ the unstable

1280: solutions belong to family  $\omega_c$. It is shown that this family

1281: contents the solutions responsible for the pion condensation.

1282: 

1283: It is possible to expect that a new state of medium after  phase

1284: transition is different depending the value of $G'$. We investigated

1285: the dependence of the maximal growth rate of perturbations

1286: at several $G'$ in the vicinity of $G'_{tr}$ on density. But this

1287: dependence does not exhibit the noticeable modifications at

1288: $G'_{tr}$.

1289: 

1290: 

1291: Author thanks S.~V. Tolokonnikov for the constructive

1292: criticism. Also, author thanks V.~R.~Shaginian and

1293: E.~G.~Drukarev

1294: and especially M.~G.~Ryskin for the numerical fruitful

1295: discussions.

1296: The work was supported by the grants RFBR-03-02-17724.

1297: 

1298: 

1299: \section{ Appendix A.}

1300: 

1301: In this appendix the scalar part of the pion self-energy,

1302: $\Pi_S$, (2) and its influence on the behavior of the curves in

1303: Fig.~1 are considered.  We construct $\Pi_S$ using the relation

1304: of Gell-Mann--Oakes--Renner \cite{DRS}, \cite{GMOR}. It permits

1305: to express the pion mass in medium through the quark condensates

1306: in nuclear matter

1307: $$

1308:  \label{a1} m^{*2}_\pi\ =\ -\frac{\langle NM|\bar

1309: qq|NM\rangle(m_u+m_d)}{ 2f^{*2}_\pi}\ , \eqno{(A1)}

1310: $$

1311: where $m_u,m_d$ - masses of

1312:  $u$- and $d$-quarks ($m_u+m_d$=11~MeV),

1313: $f^*_\pi$ - coupling constant of the pion decay in the medium

1314: (here we assume that $f^*_\pi=f_\pi$=92~MeV, i.e., the vacuum and

1315: medium values of $f_\pi$ coincide).

1316: The value $\kappa=\langle NM|\bar qq|NM\rangle$ is a scalar

1317: quark condensate in nuclear matter. It may be present as

1318: $$

1319: \kappa\ =\ \kappa_0+\rho\langle N|\bar qq|N\rangle +...

1320: \eqno{(A2)}

1321: $$

1322: Here $\kappa_0$ is a scalar quark condensate in vacuum,

1323: $\kappa_0$=-0.03~GeV$^3$; $\langle N|\bar qq|N\rangle$ is the

1324: matrix element of the scalar quark condensate calculated over

1325: the nucleon $\langle N|\bar qq|N\rangle\simeq 8$. Then from (A1)

1326: we obtain

1327: $$

1328: m^{*2}_\pi\ =\ m^2_\pi-\rho\,\frac{\langle N|\bar

1329: qq|N\rangle (m_u+m_d)}{2f^2_\pi}\ .

1330: \eqno{(A3)}

1331: $$

1332: On the other hand, we can define $m^{*2}_\pi$ using Eq.~(1) as

1333: $$

1334: m^{*2}_\pi\ =\ m^2_\pi+\Pi(\omega,k=0)\ .

1335: \eqno{(A4)}

1336: $$

1337: 

1338: The form of $\Pi_S$ can be obtained comparing (A2) and (A3)

1339: provided that $\Pi_P(\omega,k=0)=0$

1340: $$

1341: \Pi_S\ =\ -\rho\ \frac{\langle N|\bar qq|N\rangle(m_u+m_d)}{

1342: 2f^2_\pi}\ .

1343: \eqno{(A5)}

1344: $$

1345: 

1346: In Fig.~1 the dotted lines correspond to the critical density

1347: $\rho_c$ and wave vector $k_c$ obtained for Eq.~(1) with the

1348: pion self-energy (\ref{1f}). (Recall that

1349: for each $G'$ in Fig.~1 we show the minimal $\rho_c$ when

1350: solution $\omega$=0 appears).  In this model the transition

1351: value of  $G'$ is $G'_{tr}$=0.0026.

1352: 

1353: \section{Appendix B}

1354: Here we construct the unphysical sheets $\tilde I$ and $\tilde I'$.

1355: The solutions of Eq.~(19) that are presented by  the curve

1356: $ab$ in Fig.~2 are placed on these sheets.

1357: 

1358: To pass on to the sheet $\tilde I$ we do the analytical continuation

1359: in $\omega$ of

1360: $ln(z)\equiv ln\left(\frac{\omega m}

1361: {\omega m-kp_F +\frac12 k^2}\right)$

1362: from the lower edge of the cut $I$ upwards, adding $2\pi i$ to

1363: $ln (z)$.

1364: The sheet $\tilde I$ is situated above the physical sheet.

1365: The value of $ln\left(\frac{\omega m}

1366: {\omega m-kp_F +\frac12 k^2}\right)$  above the upper edge of  the cut

1367: $I$ is $ln(-z)-\pi i+2\pi i$. This expression is continued on the sheet

1368: $\tilde I$.

1369: 

1370: In similar way we construct the sheet $\tilde I'$ making the

1371: analytical continuation of

1372: $ln(z')\equiv ln\left(\frac

1373: {\omega m+kp_F -\frac12 k^2}{-\omega m}\right)$

1374: from the lower edge of cut $I'$, adding $2\pi i$ to logarithm

1375: values.

1376: 

1377: 

1378: \newpage

1379: 

1380: \begin{thebibliography}{99}

1381: \bibitem{Mi}

1382:  A.~B. Migdal, Rev. Mod. Phys.  {\bf 50}, 107 (1978);

1383: A.~B.~Migdal, D.~N.~Voskresenskii, E.~E.~Saperstein, and

1384: M.~A.~Troitskii, Phys. Rep. {\bf 192}, 179 (1990).

1385: 

1386: \bibitem{DRS} E.~G.~Drukarev, M.~G.~Ryskin, and V.~A.~Sadovnikova,

1387: Eur. Phys. J.  {\bf A4}, 171 (1999).

1388: 

1389: \bibitem{SR}  V.~A.~Sadovnikova and M.~G.~Ryskin, Phys. At. Nucl. {\bf

1390: 64}, 440 (2001).

1391: 

1392: \bibitem{SS}

1393: R.~F.~Sawyer and D.~J. Scalapino,  Phys. Rev. {\bf D7}, 953 (1973).

1394: 

1395: \bibitem{BW} G.~E. Brown and W.~Weise, Phys. Rep. {\bf 27}, 1 (1976).

1396: 

1397: \bibitem{OTW} E. Oset, W. Weise, and H. Toki, Phys. Rep. {\bf 83},

1398: 281 (1982).

1399: 

1400: \bibitem{MSG} G. Mao, L. Neise, H. Stocker, and W. Greiner, Phys. Rev.

1401: {\bf C59}, 1674 (1999).

1402: 

1403: \bibitem{HR} J. Helgesson and J. Randrap, Phys. Rev, {\bf C52}, 427

1404: (1995).

1405: 

1406: \bibitem{GK} C. Gale and J.~Kapusta, Phys. Rev. {\bf C35}, 2107 (1987).

1407: 

1408: \bibitem{XX} L.~H. Xia, C.M. Ko, L. Xiong, and J.~Q. Wu, Nucl. Phys. {\bf

1409: A485}, 721 (1988).

1410: 

1411: 

1412: \bibitem{LN} L. Liu and M. Nakano, Nucl. Phys. {\bf A618}, 337 (1997).

1413: 

1414: \bibitem{AM} J. Diaz Alonso and A. Perez~Canyellas, Nucl. Phys. {\bf

1415: A526}, 623 (1991); L.~Mornas, nucl-th/0107005.

1416: 

1417: \bibitem{LP} L.~B.~Leinson and A.~Perez, Eur.~Phys.~J. {\bf A20}, 371

1418: (2004).

1419: 

1420: \bibitem{Lutz} M.~F.~M.~Lutz, Phys.~Lett. {\bf B552}, 159 (2003);

1421: Phys.Lett.  {\bf B 566}, 277 (2003).

1422: 

1423: \bibitem{BBN} S.-O. Backman, G.~E. Brown, and J.~A.~Niskanen, Phys.~Rep. {\bf

1424: 124}, 1 (1985).

1425: 

1426: \bibitem{GOS} C. Garsia-Recio, E. Oset, and L.~L.~Salsedo, Phys.~Rev. {\bf

1427: C37}, 194 (1988).

1428: 

1429: \bibitem{CE} G. Chanfray and M. Ericson, Phys.~Lett. {\bf B141}, 163

1430: (1984).

1431: 

1432: \bibitem{DG} J. Delorme and P.~A.~M. Guichon, Phys.~Lett. {\bf B263},

1433: 157 (1991).

1434: 

1435: \bibitem{CC} M. Colonna and Ph. Chomaz, Phys.~Rev. {\bf C49},

1436: 1908 (1994).

1437: \bibitem{PR} C.~J.~Pethick and D.~G. Ravenhall, Ann.~Phys. {\bf 183},

1438: 131 (1988).

1439: 

1440: \bibitem{EW} T. Ericson and W. Weise, {\em``Pions and nuclei"},

1441: Clarendon Press, 1988.

1442: 

1443: \bibitem{LaLi} L.~D. Landau and E.~M. Lifshitz,  {\em``Statistical

1444: physics"} (Pergamon Press, 1989).

1445: 

1446: \bibitem{Pom} I.~Ya. Pomeranchuk, Sov. Phys. JETP {\bf 8}, 361 (1958).

1447: 

1448: 

1449: 

1450: \bibitem{PN} D. Pines and P. Nozieres,

1451: {\em``The Theory of Quantum Liquids"} (W.A.Benjamin,Inc., 1966).

1452: 

1453: \bibitem{MS} H. Mueller and B.D. Serot , Phys.~Rev. {\bf C52},

1454:  2072 (1995).

1455: 

1456: \bibitem{SPWW} J. Speth and E. Werner, W. Wild, Phys.~Rep. {\bf

1457: 33}, 127 (1977).

1458: 

1459: \bibitem{FW} A.~L. Fetter and J.~D. Walecka,{\em``Quantum theory of

1460: many-particle systems"}(Mc-Graw-Hill, New York, 1971).

1461: 

1462: 

1463: \bibitem{DM} W.H. Dickhoff, A.~Faessler, J.~Meyr-ter-Vehn,

1464: and H.~Muthner, Phys.~Rev. {\bf C23}, 1154 (1981).

1465: 

1466: \bibitem{BaW} S.-O. Backmann and W.~Weise, {\em ``Mesons in nuclei''},

1467: v.III, 1095, Ed. by M.~Rho, D.~Wilkinson, (North-Holland Pub.

1468: Company, 1976).

1469: 

1470: 

1471: \bibitem{BR} G.~E.~Brown and M.~Rho, Phys.~Rev.~Lett. {\bf 66}, 2720

1472: (1991).

1473: 

1474: \bibitem{MRL} R.~J. Furnstahl, D.~K.~Griegel, and T~.D.~Cohen,

1475: Phys.~Rev.  {\bf C46}, 1507 (1992).

1476: 

1477: 

1478: \bibitem{DRSF}  E.~G. Drukarev,  M.~G. Ryskin,  V.~A.~Sadovnikova,

1479: Th.~Gutsche, and A.~Faessler, Phys.~Rev. {\bf C69}, 065210 (2004).

1480: 

1481: \bibitem{GMOR} M. Gell-Mann, R. Oakes, and B. Renner, Phys.~Rev.

1482: {\bf 175}, 2195 (1968).

1483: 

1484: \end{thebibliography}

1485: 

1486: \newpage

1487: 

1488: \section{Figure captions}

1489: \noindent

1490: FIG.~1.

1491: Critical values of nuclear density $\rho_c$ and  wave

1492: vector $k_c$, at which in Eq.~(1) solutions $\omega$=0 appear.

1493: Results for three models of the pion self-energy are

1494: presented:  solid lines correspond

1495: $\Pi=\Pi_N + \Pi_\Delta = \Pi_P$ (\ref{1f}), (\ref{2f});

1496: dashed curves are  for  $\Pi= \Pi_N$;

1497:  dotted curves are for the total $\Pi= \Pi_P+\Pi_S$

1498: (\ref{1f}).

1499: 

1500: \noindent

1501: FIG.~2. Complex plane of the frequency  $\omega$.

1502: a) Numbers $I$ and $II$ denote the  cuts (\ref{12})  of

1503: $\phi(\omega,k)$ and $I'$ and $II'$ denote the  cuts of

1504: $\phi(-\omega,k)$;

1505: b) Unphysical sheets are shown:

1506:  sheet $I$ (the shading with the right  slope)  and

1507:  sheet $I'$ (the shading with the left  slope).

1508: 

1509: \noindent

1510: FIG.~3. Solutions of Eqs.~(\ref{6x}),~(\ref{9x}),~(\ref{13})

1511: at $G'<0$ in the kinetic theory of Landau. The shading with the

1512: left  slope marks unphysical sheet $I$ and the horizontal shading

1513: refers to $\tilde I$.

1514: Results are obtained at  $\rho=\rho_0$.

1515: 

1516: \noindent

1517: FIG.~4. Imaginary solutions $\omega_{sd}(k)$ to

1518: Eq.~(\ref{17x}) obtained in RPA for  following values of

1519:  $G'$:  (1) -0.4; (2) -0.9; (3) -1.02; (4) -1.1; (5) -1.2.

1520: Results are obtained at  $\rho=\rho_0$.

1521: 

1522: \noindent

1523: FIG.~5.

1524: Imaginary solutions $\omega_{sd}(k)$ of  Eq.~(\ref{7}). 

1525: a)~Dependence of $\omega_{sd}(k)$  on  $f_{\pi NN}$:

1526: the curve (1) $f_{\pi NN}$=0; (2) 0.4; (3) 0.6; (4) 1.0.

1527: The results are obtained at  $\rho=\rho_0$ and

1528: $G'_{NN}$=-0.4.

1529: b)~Dependence of $\omega_{sd}(k)$ on the value of density: the curve (1)

1530: $\rho/\rho_0$=1; (2) 0.318; (3) 0.265; (4) 0.221.

1531: The results are obtained at  $f_{\pi NN}$=1 and $G'_{NN}$=-0.4.

1532: Point $A$ marks solution with $\omega$=0.

1533: 

1534: \noindent

1535: FIG.~6. Different types of solutions of Eq.~(1):

1536: spin-isospin zero-sound branch

1537: $\omega^\pi_s(k)$, pion branch $\omega_\pi(k)$, isobar branch

1538: $\omega_\Delta(k)$ and condensate branch $\omega_c(k)$.

1539: Results are obtained at $\frac{\rho}{\rho_0}$=1.78 and $G'_{NN}$=2.

1540: We computed that  $k_1$=1.35$m_\pi$, $k_2$=2.35$m_\pi$.

1541: 

1542: \noindent

1543: FIG.~7. Branches  $\omega_c(k)$ and $\omega'_c(k)$ on the complex

1544: plane of frequency.

1545: The shading with the right (left) slope marks the unphysical

1546: sheet $I$ ($I'$).

1547: The physical sheet is above the abscissas axis. Axes of

1548: reference are shown by the dotted lines. The solid curve

1549: correspond to $\omega_c(k)$ and the dashed one is for

1550: $\omega'_c(k)$.  The number `1' marks  start of branches

1551: at $k$=0; `2' is the coalescence point at

1552: $k_m$=1.28$m_\pi$; `3' is the turning point at $k_t$=

1553: 1.88$m_\pi$.  Variables $k_1, k_2$, $\frac{\rho}{\rho_0}$

1554: and $G'_{NN}$ are the same as in Fig.~6.

1555: 

1556: \noindent

1557: FIG.~8. Branches  $\omega_c(k)$ and $\omega'_c(k)$ on the

1558: complex plane of frequency at  different densities

1559: $\frac{\rho}{\rho_0}$.  In  figures b,c,d  branches are

1560: shown at $k\leq k_t$. Point $B$ marks  solution with

1561: $\omega$=0.  Notations of lines and sheets are the same as in

1562: Figs.~6,~7.

1563: 

1564: 

1565: \noindent

1566: FIG.~9. Transition of a zero solution from  family

1567: $\omega_{sd}$ to  family  $\omega_c$. The solid, dashed

1568: and dotted curves  denote $\omega_c(k)$, $\omega'_c(k)$

1569: and $\omega^\pi_{sd}(k)$, correspondingly.

1570: a) $G'$=-0.0020. The number `1' corresponds to $k$=0; `2' means

1571:  turning point of $\omega^\pi_{sd}(k)$:

1572: $k_t$=0.077$m_\pi$; `3' $k$=0.1$m_\pi$. Arrows show 

1573: movement of  branches with $k$. Point $C$ corresponds to

1574: $\omega^\pi_{sd}(k_c)$=0.

1575: b) $G'$=-0.0018. Branches are shown for $k\leq 0.094m_\pi$ (point

1576: `2').

1577: c) Continuation of figure b. Branches are shown for

1578: $k>0.094m_\pi$.  Point `3' marks the coalescence point of

1579: $\omega_c(k)$ and  $\omega'_c(k)$ at $k$=0.0962$m_\pi$.

1580: Point `4' shows the coalescence point of

1581: $\omega'_c(k)$ and $\omega^\pi_{sd}(k)$  at $k$=0.109$m_\pi$.

1582: Point $D$ corresponds to $\omega_c(k_c)$=0.

1583: 

1584: 

1585: \noindent

1586: FIG.~10. Dependence of the maximal growth rate of

1587: fluctuations on density.  The curve (1) corresponds to

1588: $G'=G'_{tr}$=-0.0019; (2) $G'=G'_{tr}$-0.02; (3)

1589: $G'=G'_{tr}$+0.02; (4) $G'=G'_{tr}$-0.2; (5) $G'=G'_{tr}$+0.2.

1590: 

1591: \newpage

1592: 

1593: %Fig.1

1594: \begin{figure}

1595: \centering{\epsfig{figure=f1.eps,width=8cm}}

1596: \caption{}

1597: \end{figure}

1598: 

1599: %Fig.2

1600: \begin{figure}

1601: \centering{

1602: \epsfig{figure=f2a.eps,width=8cm}

1603: \epsfig{figure=f2b.eps,width=8cm}

1604: }

1605: \caption{}

1606: \end{figure}

1607: 

1608: %Fig.3

1609: \begin{figure}

1610: \centering{\epsfig{figure=f3.eps,width=8cm}}

1611: \caption{}

1612: \end{figure}

1613: 

1614: %Fig.4

1615: \begin{figure}

1616: \centering{\epsfig{figure=f4a.eps,width=9cm}}

1617: \caption{}

1618: \end{figure}

1619: 

1620: %Fig.5

1621: \begin{figure}

1622: \centering{\epsfig{figure=f5a.eps,width=8cm}}

1623: \centering{\epsfig{figure=f5b.eps,width=8cm}}

1624: \caption{}

1625: \end{figure}

1626: 

1627: %Fig.6

1628: \begin{figure}

1629: \centering{\epsfig{figure=f6.eps,width=8cm}}

1630: \caption{}

1631: \end{figure}

1632: 

1633: %Fig.7

1634: \begin{figure}

1635: \centering{\epsfig{figure=f8.eps,width=9cm}}

1636: \caption{}

1637: \end{figure}

1638: 

1639: 

1640: %Fig.8

1641: \begin{figure}

1642: \centering{\epsfig{figure=f7ab.eps,width=12cm}}

1643: \centering{\epsfig{figure=f7cd.eps,width=12cm}}

1644: \caption{}

1645: \end{figure}

1646: 

1647: %Fig.9

1648: \begin{figure}

1649: \centering{\epsfig{figure=f01.eps,width=7cm}}

1650: \centering{\epsfig{figure=f09.eps,width=14cm}}

1651: \caption{}

1652: \end{figure}

1653: 

1654: %Fig.10

1655: \begin{figure}

1656: \centering{\epsfig{figure=f10.eps,width=7cm}}

1657: \caption{}

1658: \end{figure}

1659: 

1660: 

1661: \end{document}

1662: 

1663: