1: %

2: % Preon stars

3: %

4: % Johan Hansson' & Fredrik Sandin"

5: % Luleå University of Technology

6: % SE-971 87 Luleå, Sweden

7: %

8: % ' E-mail: hansson@mt.luth.se

9: % " E-mail: fredrik.sandin@mt.luth.se

10: %

11: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

12: \documentclass[showpacs,11pt]{revtex4}

13: 

14: \usepackage{epsfig}

15: \usepackage{amssymb}

16: 

17: \newcommand{\figurewidth}{6in}

18: 

19: \begin{document}

20: 

21: \title{Preon stars: a new class of cosmic compact objects}

22: 

23: \author{J. Hansson\footnote{e-mail: \tt c.johan.hansson@ltu.se} \& F. Sandin\footnote{e-mail: \tt fredrik.sandin@ltu.se}}

24: 

25: \affiliation{Department of Physics, Lule{\aa} University of Technology, SE-971 87 Lule\aa, Sweden}

26: 

27: \begin{abstract}

28: In the context of the standard model of particle physics, there is a definite upper limit

29: to the density of stable compact stars. However, if a more fundamental level of elementary

30: particles exists, in the form of preons, stability may be re-established beyond this limiting

31: density. We show that a degenerate gas of interacting fermionic preons does allow for stable

32: compact stars, with densities far beyond that in neutron stars and quark stars. In keeping with

33: tradition, we call these objects ``preon stars", even though they are small and light compared

34: to white dwarfs and neutron stars. We briefly note the potential importance of preon stars in

35: astrophysics, {\it e.g.}, as a candidate for cold dark matter and sources of ultra-high energy

36: cosmic rays, and a means for observing them.

37: \end{abstract}

38: 

39: \pacs{12.60.Rc - 04.40.Dg - 97.60.-s - 95.35.+d}

40: 

41: \maketitle

42: 

43: \section{Introduction}

44: 

45: The three different types of compact objects traditionally considered in astrophysics

46: are white dwarfs, neutron stars (including quark and hybrid stars), and black holes.

47: The first two classes are supported by Fermi pressure from their constituent particles.

48: For white dwarfs, electrons provide the pressure counterbalancing gravity. In neutron

49: stars, the neutrons play this role. For black holes, the degeneracy pressure is overcome

50: by gravity and the object collapses indefinitely, or at least to the Planck density.

51: 

52: The distinct classes of degenerate compact stars originate directly from the properties

53: of gravity, as was made clear by a theorem of Wheeler and collaborators in the mid

54: 1960s~\cite{Harrison}. The theorem states that for the solutions to the stellar structure

55: equations, whether Newtonian or relativistic, there is a change in stability of one radial

56: mode of normal vibration whenever the mass reaches a local maximum or minimum as a function

57: of central density. The theorem assures that distinct classes of stars, such as white

58: dwarfs and neutron stars, are separated in central density by a region in which there

59: are no stable configurations.

60: 

61: In the standard model of particle physics (SM), the theory of the strong interaction

62: between quarks and gluons predicts that with increasing energy and density, the coupling

63: between quarks asymptotically fades away~\cite{Gross,Politzer}. As a consequence of this

64: ``asymptotic freedom", matter is expected to behave as a gas of free fermions at

65: sufficiently high densities. This puts a definite upper limit to the density of stable

66: compact stars, since the solutions to the stellar equations end up in a never-ending

67: sequence of unstable configurations, with increasing central density. Thus, in the

68: light of the standard model, the densest stars likely to exist are neutron stars, quark

69: stars or the potentially more dense hybrid stars~\cite{Gerlach,Glendenning2,Schertler}.

70: However, if there is a deeper layer of constituents, below that of quarks and

71: leptons, asymptotic freedom will break down at sufficiently high densities, as

72: the quark matter phase dissolves into the preon sub-constituent phase.

73: 

74: There is a general consensus among the particle physics community, that something new

75: should appear at an energy-scale of around one TeV. The possibilities are, {\it e.g.},

76: supersymmetric particles, new dimensions and compositeness. In this letter we consider

77: ``preon models"~\cite{dSouza,PreonTrinity}, \textit{i.e.}, models in which quarks and

78: leptons, and sometimes some of the gauge bosons, are composite particles built out of more

79: elementary preons. If fermionic preons exist, it seems reasonable that a new type of

80: astrophysical compact object, a {\it preon star}, could exist. The density in preon stars

81: should far exceed that inside neutron stars, since the density of preon matter must be

82: much higher than the density of nuclear and deconfined quark matter. The sequence of

83: compact objects, in order of increasing compactness, would thus be: white dwarfs,

84: neutron stars, preon stars and black holes.

85: 

86: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

87: 

88: \section{Mass-radius relations}

89: 

90: Assuming that a compact star is composed of non-interacting fermions with mass $m_f$, the

91: non-general relativistic (Chandrasekhar) expression for the maximum mass is~\cite{Chandra,Landau}:

92: \begin{equation}

93: M \simeq \frac{1}{m_f^2}\left(\frac{\hbar c}{G}\right)^{3/2}.

94: \label{chandra}

95: \end{equation}

96: This expression gives a correct order of magnitude estimate for the mass of a white dwarf and a

97: neutron star. For quark stars, this estimate cannot be used literally, since the mass of quarks

98: cannot be defined in a similar way as for electrons and neutrons. However, making the simplifying

99: assumption that quarks are massless and subject to a `bag constant', a

100: maximum mass relation can be derived~\cite{Shibaji}. The bag constant is a phenomenological

101: parameter. It represents the strong interactions that, in addition to the quark momenta,

102: contribute mass-energy to deconfined quark matter, {\it i.e.}, in the same way as the bag constant

103: for ordinary hadrons~\cite{bag}. The result in~\cite{Shibaji} is somewhat similar to the

104: Chandrasekhar expression, but the role of the fermion mass is replaced by the bag constant $B$:

105: \begin{eqnarray}

106: M &=& \frac{16 \pi B R^3}{3 c^2}, \label{mchandra} \\

107: R &=& \frac{3 c^2}{16 \sqrt{\pi G B}}. \label{rchandra}

108: \end{eqnarray}

109: 

110: For preon stars, one can naively insert a preon mass of $m_f \simeq 1$~TeV/c$^2$ in eq.~(\ref{chandra})

111: to obtain a preon star mass of approximately one Earth mass (M$_\oplus \simeq 6\times10^{24}$~kg). However,

112: the energy scale of one TeV should rather be interpreted as a length scale, since it originates from the

113: fact that in particle physics experiments, no substructure has been found down to a scale of a few hundred

114: GeV ($\hbar c/$GeV$\,\simeq10^{-18}$~m). Since preons must be able to give light particles, {\it e.g.},

115: neutrinos and electrons, the ``bare" preon mass presumably is fairly small and a large fraction of the

116: mass-energy should be due to interactions. This is the case for deconfined quark matter, where the bag

117: constant contributes more than $10$\% of the energy density. Guided by this observation, and lacking

118: a quantitative theory for preon interactions, we assume that the mass-energy contribution from preon

119: interactions can be accounted for by a bag constant. We estimate the order of magnitude for the preon

120: bag constant by fitting it to the minimum density of a composite electron, with mass $m_e=511$~keV/c$^2$

121: and ``radius" {$R_{e} \lesssim \hbar c/$TeV$\,\simeq 10^{-19}$~m}. The bag-energy is roughly

122: $4 B\langle V\rangle$~\cite{bag}, where $\langle V\rangle$ is the time-averaged volume of the bag

123: (electron), so the bag constant is:

124: \begin{equation}

125: 	B \simeq \frac{E}{4\langle V\rangle} \gtrsim \frac{3\times 511\,{\rm keV}}{16\pi(10^{-19}\,{\rm m})^3}

126: 	\simeq 10^{4}\,{\rm TeV/fm}^{3} 

127: 	\Longrightarrow\, B^{1/4} \gtrsim 10\,{\rm GeV}.

128: \end{equation}

129: Inserting this value of $B$ in eqs.~{(\ref{mchandra}),~(\ref{rchandra})}, we obtain an estimate

130: for the maximum mass, $M_{max} \simeq 10^2$~M$_\oplus$, and radius, $R_{max} \simeq 1$~m, of a

131: preon star.

132: 

133: Since $B^{1/4}\simeq 10$~GeV only is an order of magnitude estimate for the minimum value of $B$,

134: in the following, we consider the bag constant as a free parameter of the model, with a lower limit

135: of $B^{1/4}=10$~GeV and an upper limit chosen as $B^{1/4}=1$~TeV.

136: The latter value corresponds to an electron ``radius" of $\hbar c/10^3\,$TeV$\,\simeq 10^{-22}$~m.

137: In figs.~\ref{maxmass}~and~\ref{maxr} the (Chandrasekhar) maximum mass and radius of a preon

138: star are plotted as a function of the bag constant.

139: 

140: \begin{figure}

141: \epsfig{file=fig1.eps,width=\figurewidth,clip=}

142: \caption{The maximum mass and corresponding central density $\rho_c$ of a preon star vs. the bag

143: constant~$B$. The solid lines represent the general relativistic OV solutions, while the dashed

144: line represents the Newtonian (Chandrasekhar) estimate. Despite the high central density, the mass

145: of these objects is below the Schwarzschild limit, as is always the case for static solutions to

146: the stellar equations. $M_\oplus\simeq6\times10^{24}$~kg is the Earth mass.}

147: \label{maxmass}

148: \end{figure}

149: 

150: Due to the extreme density of preon stars, a general relativistic treatment is necessary.

151: This is especially important for the analysis of stability when a preon star is subject to

152: small radial vibrations. In this introductory article we will neglect the effects of rotation

153: on the composition. Thus, we can use the Oppenheimer-Volkoff (OV) equations~\cite{OV}

154: for hydrostatic, spherically symmetric equilibrium:

155: \begin{eqnarray}

156: \frac{dp}{dr} &=& -\frac{G\left(p+\rho c^2\right)\left(m c^2 + 4 \pi r^3 p\right)}

157: 	{r \left(r c^4 - 2 G m c^2 \right)},  \\

158: \frac{dm}{dr} &=& 4 \pi r^2 \rho.\label{oveq2}

159: \end{eqnarray}

160: Here $p$ is the pressure, $\rho$ the total density and $m=m(r)$ the mass within the radial

161: coordinate $r$. The total mass of a preon star is $M = m(R)$,

162: %\begin{equation}

163: %M = 4\pi \int_0^R r^2 \rho \, dr,

164: %\end{equation}

165: where $R$ is the coordinate radius of the star. Combined with an equation of state (EoS), $p = p(\rho)$,

166: obtained from some microscopic theory, the OV solutions give the possible equilibrium states

167: of preon stars.

168: 

169: Since no theory for the interaction between preons yet exists, we make a simple assumption for the

170: EoS. The EoS for a gas of massless fermions is $\rho c^2 = 3 p$ (see, {\it e.g.},~\cite{Glendenning}),

171: independently of the degeneracy factor of the fermions. By adding a bag constant $B$, one obtains

172: $\rho c^2 = 3 p + 4 B$. This is the EoS that we have used when solving the OV equations. The obtained

173: (OV) maximum mass and radius configurations are also plotted in figs.~\ref{maxmass}~and~\ref{maxr}.

174: 

175: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

176: 

177: \section{Stability analysis}

178: 

179: A necessary, but not sufficient, condition for stability of a compact star is that

180: the total mass is an increasing function of the central density ${dM/d\rho_c>0}$~\cite{Glendenning}.

181: This condition implies that a slight compression or expansion of a star will result

182: in a less favourable state, with higher total energy. Obviously, this is a necessary

183: condition for a stable equilibrium configuration. Equally important, a star must be

184: stable when subject to radial oscillations. Otherwise, any small perturbation would

185: bring about a collapse of the star.

186: 

187: \begin{figure}

188: \epsfig{file=fig2.eps,width=\figurewidth,clip=}

189: \caption{The maximum radius and the corresponding first three eigenmode oscillation frequencies

190: ($f_0$,~$f_1$,~$f_2$) vs. the bag constant. The solid line in the left-hand picture is the ``apparent"

191: radius, $R^\infty = R / \sqrt{1-2 G M / Rc^2}$, as seen by a distant observer. The dashed line

192: represents the general relativistic coordinate radius obtained from the OV solution, while the

193: dotted line represents the Newtonian (Chandrasekhar) estimate. Since the fundamental mode $f_0$

194: is real ($\omega_0^2 > 0$), preon stars with mass below the maximum are stable, for each value of B.}

195: \label{maxr}

196: \end{figure}

197: 

198: The equations for the analysis of such radial modes of oscillation are due to

199: Chandrasekhar~\cite{Chandra2}. An overview of the theory, and some applications, can

200: be found in~\cite{Misner}. For clarity, we reproduce some of the important points.

201: Starting with the metric of a spherically symmetric equilibrium stellar model,

202: \begin{equation}

203: ds^2 = -e^{2 \nu(r)}dt^2 + e^{2 \lambda(r)}dr^2 + r^2\left(d\theta^2 + \sin^2(\theta)d\phi^2\right),

204: \end{equation}

205: and the energy-momentum tensor of a perfect fluid,

206: %\begin{equation}

207: $T_{\mu \nu}=(\rho + p)u_\mu u_\nu + p g_{\mu\nu}$,

208: %\end{equation}

209: the equation governing radial adiabatic oscillations can be derived from Einstein's equation.

210: By making an ansatz for the time dependence of the radial displacement of fluid elements

211: of the form:

212: \begin{equation}

213: \delta r(r,t) = r^{-2} e^{\nu(r)} \zeta(r) e^{i \omega t},

214: \label{displacement}

215: \end{equation}

216: the equation simplifies to a Sturm-Liouville eigenvalue equation for the eigenmodes~\cite{Chandra2,Misner}:

217: \begin{equation}

218: \frac{d}{dr}\left( P \frac{d\zeta}{dr}\right) + (Q + \omega^2 W) \zeta = 0.

219: \label{sturmeq}

220: \end{equation}

221: The coefficients $P$, $Q$ and $W$ are~\cite{Misner} (in geometric units where $G=c=1$):

222: \begin{eqnarray}

223: P &=& \Gamma r^{-2} p\,e^{\lambda(r) + 3\nu(r)}, \\

224: Q &=& e^{\lambda(r) + 3\nu(r)} \left[ (p+\rho)^{-1} r^{-2} \left(\frac{dp}{dr}\right)^2

225: 	 - 4 r^{-3}\frac{dp}{dr} - 8\pi r^{-2} p(p+\rho) e^{2 \lambda(r)} \right], \\

226: W &=& (p+\rho) r^{-2} e^{ 3\lambda(r) + \nu(r)},

227: \end{eqnarray}

228: where the adiabatic index $\Gamma$ is:

229: \begin{equation}

230: \Gamma = \frac{p+\rho}{p}\left(\frac{\partial p}{\partial \rho}\right)_S.

231: \end{equation}

232: The boundary conditions for $\zeta(r)$ are that $\zeta(r)/r^3$ is finite or zero as $r\rightarrow 0$,

233: and that the Lagrangian variation of the pressure,

234: \begin{equation}

235: \Delta p = -\frac{\Gamma p\,e^{\nu}}{r^{2}} \frac{d\zeta}{dr},

236: \end{equation}

237: vanishes at the surface of the star.

238: 

239: A catalogue of various numerical methods for the solution of eq.~(\ref{sturmeq}) can

240: be found in~\cite{Bardeen}. In principle, we first solve the OV equations, thereby

241: obtaining the metric functions $\lambda(r)$ and $\nu(r)$, as well as $p(r)$, $\rho(r)$

242: and $m(r)$. Then, the metric functions $\lambda(r)$ and $\nu(r)$ must be corrected for,

243: so that they match the Schwarzschild metric at the surface of the star (see, {\it e.g.},~\cite{Glendenning}).

244: Once these quantities are known, eq.~(\ref{sturmeq}) can be solved for $\zeta(r)$ and

245: $\omega^2$ by a method commonly known as the ``shooting" method. One starts with an

246: initial guess on $\omega^2$, and integrates eq.~(\ref{sturmeq}) from $r=0$ to the

247: surface of the star. At this point $\zeta(r)$ is known, and $\Delta p$ can be calculated.

248: The number of nodes of $\zeta(r)$ is a non-decreasing function of $\omega^2$ (due to

249: Sturm's oscillation theorem). Thus, one can continue making educated guesses for

250: $\omega^2$, until the correct boundary condition ($\Delta p = 0$) and number of nodes

251: are obtained. This method is simple to use when only a few eigenmodes are needed.

252: 

253: Due to the time dependence in eq.~(\ref{displacement}), a necessary (and sufficient) condition

254: for stability is that all $\omega_i^2$ are positive. Since $\omega_i^2$ are eigenvalues of a

255: Sturm-Liouville equation, and governed by Sturm's oscillation theorem, it is sufficient to prove

256: that the fundamental mode, $\omega^2_0$, is greater than zero for a star to be stable.

257: In fig.~\ref{b100gev} the first three oscillation frequencies, $f_i=\omega_i/2\pi$, for various

258: stellar configurations with $B^{1/4}=100$~GeV are plotted.

259: %

260: \begin{figure}

261: \epsfig{file=fig3.eps,width=\figurewidth,clip=}

262: \caption{The mass and the first three eigenmode oscillation frequencies ($f_0$,~$f_1$,~$f_2$) vs. the

263: central density $\rho_c$ of preon stars. Here, a fixed value of $B^{1/4}=100$~GeV has been used. For the

264: maximum mass configuration, the fundamental mode $f_0$ has zero frequency, indicating the onset of

265: instability. Preon stars with mass and density below the maximum mass configuration of this sequence

266: are stable.}

267: \label{b100gev}

268: \end{figure}

269: %

270: In agreement with the turning point theorem of Wheeler {\it et al.}~\cite{Harrison},

271: the onset of instability is the point of maximum mass, as $\omega_0^2$ becomes negative for

272: higher central densities. Thus, for this value of the bag constant, preon stars are stable

273: up to the maximum mass configuration.

274: In order to see if the same is true for other values of $B$, we plot the first three

275: oscillation frequencies as a function of $B$, choosing the maximum radius configuration

276: for each $B$. The result can be found in fig.~\ref{maxr}. Indeed, the previous

277: result is confirmed; all configurations up to the maximum mass are stable.

278: 

279: The eigenmode frequencies for radial oscillations of preon stars are about six orders

280: of magnitude higher than for neutron stars. This result can also be obtained by making

281: a simple estimate for the frequency of the fundamental mode. The radius of a preon star

282: is a factor of $\sim 10^5$ smaller than neutron stars. Hence, if the speed of sound is

283: similar in preon stars and neutron stars, the frequency would increase by a factor of

284: $\sim 10^5$, giving GHz frequencies. If the speed of sound is higher in preon stars,

285: say approaching the speed of light, the maximum frequency is

286: $\sim 10^8$~m~s$^{-1}/\,0.1$~m~$\simeq 1$~GHz.

287: Thus, in either case, GHz oscillation frequencies are expected for preon stars.

288: %Thus, GHz oscillation frequencies are expected for preon stars.

289: 

290: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

291: 

292: \section{Potential astrophysical consequences and detection}

293: 

294: If preon stars do exist, and are as small as $10^{-1}-10^{-4}$~m, it is plausible that primordial

295: preon stars (or ``nuggets") formed from density fluctuations in the early universe. As this material

296: did not take part in the ensuing nucleosynthesis, the abundance of preon nuggets is not constrained

297: by the hot big bang model bounds on baryonic matter. Also, preon nuggets are immune to Hawking

298: radiation~\cite{Hawking} that rapidly evaporates small primordial black holes, making it possible

299: for preon nuggets to survive to our epoch. They can therefore serve as the mysterious dark matter

300: needed in many dynamical contexts in astrophysics and cosmology~\cite{DM1,DM2}.

301: 

302: Preon stars born out of the collapse of massive ordinary stars~\cite{Hypernova} cannot contribute

303: much to cosmological dark matter, as that material originally is baryonic and thus constrained

304: by big bang nucleosynthesis. However, they could contribute to the dark matter in galaxies. Roughly

305: $4$\% of the total mass of the universe is in baryonic form \cite{MAP}, but only $0.5$\% is observed

306: as visible baryons~\cite{PartProp}. Assuming, for simplicity, that all dark matter

307: $\rho_{DM} = 10^{-25}$~g/cm$^3$ in spiral galaxies, {\it e.g.}, our own Milky Way, is in the form

308: of preon stars with mass $10^{24}$~kg, the number density of preon stars is of the order of

309: $10^4$ per cubic parsec ($1$~parsec$\,\simeq 3.1\times 10^{16}$~m). This translates into one preon

310: star per $10^6$ solar system volumes. However, even though it is not ruled out a priori, the possibility

311: to form a very small and light preon star in the collapse of a large massive star remains to be more

312: carefully investigated. In any case, preon nuggets formed in the primordial density fluctuations could

313: account for the dark matter in galaxies. The existence of such objects can in principle be tested by

314: gravitational microlensing experiments.

315: 

316: Today there is no known mechanism for the acceleration of cosmic rays with energies above

317: $\sim 10^{17}$~eV. These so-called ultra-high energy cosmic rays~\cite{UHE} (UHE CR) are

318: rare, but have been observed with energies approaching $10^{21}$~eV. The sources of UHE CR must,

319: cosmologically speaking, be nearby ($\lesssim 50$ Mpc $\simeq 150$ million light years) due to the

320: GZK-cutoff energy $\sim 10^{19}$~eV~\cite{GZK1,GZK2}, since the cosmic microwave background

321: is no longer transparent to cosmic rays at such high energies. This requirement is very puzzling,

322: as there are no known sources capable of accelerating UHE CR within this distance.

323: Preon stars open up a new possibility. It is known that neutron stars, in the form of pulsars,

324: can be a dominant source of galactic cosmic rays~\cite{Gold}, but cannot explain UHE CR. If for

325: preon stars we assume, as in models of neutron stars, that the magnetic flux of the parent star is

326: (more or less) frozen-in during collapse, induced electric fields more than sufficient for the

327: acceleration of UHE CR become possible. As an example, assume that the collapse of a massive star

328: is slightly too powerful for the core to stabilize as a $10$~ms pulsar with radius $10$~km, mass

329: $1.4 M_\odot$ ($M_\odot\simeq 2\times10^{30}$~kg is the mass of our sun) and magnetic field

330: $10^8$~T, and instead collapses to a preon star state with radius $1$~m and mass $10^2\,M_\oplus$.

331: An upper limit estimate of the induced electric field of the remaining ``preon star pulsar"

332: yields $\sim 10^{34}$~V/m, which is more than enough for the acceleration of UHE CR. Also, such strong

333: electric fields are beyond the limit where the quantum electrodynamic vacuum is expected to break

334: down, $|\textbf{E}| > 10^{18}$~V/m, and spontaneously start pair-producing particles~\cite{QED}.

335: This could provide an intrinsic source of charged particles that are accelerated by the electric

336: field, giving UHE CR. With cosmic ray detectors, like the new Pierre Auger Observatory~\cite{Auger},

337: this could provide means for locating and observing preon stars.

338: 

339: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

340: 

341: \section{Conclusions}

342: 

343: In this letter we argue that if there is a deeper layer of fermionic constituents, so-called preons,

344: below that of quarks and leptons, a new class of stable compact stars could exist. Since no detailed

345: theory yet exists for the interaction between preons, we assume that the mass-energy contribution

346: from preon interactions can be accounted for by a `bag constant'. By fitting the bag constant to the

347: energy density of a composite electron, the maximum mass for preon stars can be estimated to

348: $\sim 10^{2}$~M$_\oplus$ ($M_\oplus\simeq6\times10^{24}$~kg being the Earth mass), and their maximum

349: radius to $\sim 1$~m. The central density is at least of the order of $10^{23}$~g/cm$^3$. Preon stars

350: could have formed by primordial density fluctuations in the early universe, and in the collapse of

351: massive stars. We have briefly noted their potential importance for dark matter and ultra-high energy

352: cosmic rays, connections that also could be used to observe them. This might provide alternative means

353: for constraining and testing different preon models, in addition to direct tests~\cite{PreonTrinity}

354: performed at particle accelerators.

355: 

356: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

357: 

358: \section{Acknowledgements}

359: 

360: F. Sandin acknowledges support from the Swedish National Graduate School of Space Technology.

361: We thank S. Fredriksson for several useful discussions and for reading the manuscript.

362: 

363: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

364: 

365: \begin{thebibliography}{00}

366: 

367: \bibitem{Harrison} Harrison B.K., Thorne K.S., Wakano M. \& Wheeler J.A., \textit{Gravitation Theory and Gravitational Collapse} (University of Chicago Press, Chicago, 1965).

368: 

369: \bibitem{Gross} Gross D.J. \& Wilczek F., Phys. Rev. Lett. {\bf 30}, 1323 (1973).

370: 

371: \bibitem{Politzer} Politzer H.D., Phys. Rev. Lett. {\bf 30}, 1346 (1973).

372: 

373: \bibitem{Gerlach} Gerlach U.H., Phys. Rev. {\bf 172}, 1325 (1968), \\

374: 	Gerlach U.H., {\it A third family of stable equilibria}, Ph.D. thesis,

375: 	Princeton University, 1968.

376: 

377: \bibitem{Glendenning2} Glendenning N.K. \& Kettner C., Astron. Astrophys. {\bf 353}, L9 (2000).

378: 

379: \bibitem{Schertler} Schertler K. {\it et al.}, Nucl. Phys. {\bf A 677}, 463 (2000). % astro-ph/0001467

380: 

381: \bibitem{dSouza} D'Souza I.A. \& Kalman C.S., \textit{Preons} (World Scientific, Singapore, 1992).

382: 

383: \bibitem{PreonTrinity} Dugne J.-J., Fredriksson S. \& Hansson J., Europhys. Lett. {\bf 57}, 188 (2002).

384: 

385: \bibitem{Chandra} Chandrasekhar S., \textit{An Introduction to the Study of Stellar Structure} (Dover Publications, New York, 1958).

386: 

387: \bibitem{Landau} Landau L.D., Phys. Z. Sowjetunion {\bf 1}, 285 (1932).

388: 

389: \bibitem{Shibaji} Banerjee S., Ghosh S.K. \& Raha S., J. Phys. {\bf G 26}, L1 (2000).

390: 

391: \bibitem{bag} Chodos A. \textit{et al.}, Phys. Rev. {\bf D 9}, 3471 (1974).

392: 

393: \bibitem{OV} Oppenheimer J.R. \& Volkoff G., Phys. Rev. {\bf 55}, 374 (1939).

394: 

395: \bibitem{Glendenning} Glendenning N.K., \textit{Compact Stars} (Springer-Verlag, New York, 1997).

396: 

397: %\bibitem{Drake} Drake J.J. \textit{et al.}, Astrophys. J. {\bf 572}, 996 (2002).

398: 

399: \bibitem{Chandra2} Chandrasekhar S., Phys. Rev. Lett. {\bf 12}, 114 (1964).

400: 

401: \bibitem{Misner} Misner C.W., Thorne K.S. \& Wheeler J.A., \textit{Gravitation} (Freeman and Co., San Francisco, 1973).

402: 

403: \bibitem{Bardeen} Bardeen J.M., Thorne K.S. \& Meltzer D.W., Astrophys. J. {\bf 145}, 505 (1966).

404: 

405: \bibitem{Hawking} Hawking S.W., Commun. Math. Phys. {\bf 43}, 199 (1975).

406: 

407: \bibitem{DM1} Turner M.S., Phys. Rep. {\bf 333}, 619 (2000).

408: 

409: \bibitem{DM2} Bergstr\"{o}m L., Rep. Progr. Phys. {\bf 63}, 793 (2000).

410: 

411: \bibitem{Hypernova} Paczy\'{n}ski B., Astrophys. J. {\bf 494}, L45 (1998).

412: 

413: \bibitem{MAP} Spergel D.N. \textit{et al.}, Astrophys. J. Suppl. {\bf 148}, 175 (2003).

414: 

415: \bibitem{PartProp} Hagiwara K. \textit{et al.}, Phys. Rev. {\bf D 66}, 010001 (2002).

416: 

417: \bibitem{UHE} Nagano M. \& Watson A.A., Rev. Mod. Phys. {\bf 72}, 689 (2000).

418: 

419: \bibitem{GZK1} Greisen K., Phys. Rev. Lett. {\bf 16}, 748 (1966).

420: 

421: \bibitem{GZK2} Zatsepin G.T. \& Kuzmin V.A., JETP Lett. {\bf 41}, 78 (1966).

422: 

423: \bibitem{Gold} Gold T., Nature {\bf 221}, 25 (1969).

424: 

425: \bibitem{QED} Schwinger J., Phys. Rev. {\bf 82}, 664 (1951).

426: 	

427: \bibitem{Auger} Dova M.T. \textit{et al.}, Nucl. Phys. Proc. Suppl. {\bf 122}, 170 (2003).

428: 

429: \end{thebibliography}

430: 

431: \end{document}

432: