1: %\documentclass[12pt,preprint{aastex}

2: \documentclass{emulateapj}

3: %\usepackage{emulateapj5}

4: \usepackage{graphicx}

5: \def\gsim{\;\lower4pt\hbox{${\buildrel\displaystyle >\over\sim}$}\;}

6: \def\lsim{\;\lower4pt\hbox{${\buildrel\displaystyle <\over\sim}$}\;}

7: \def\grls{\;\lower4pt\hbox{${\buildrel\displaystyle >\over <}$}\;}

8: 

9: \begin{document}

10: \title{Magnetic Sunyaev-Zel'dovich effect in galaxy clusters}

11: %\title{Magnetic Sunyaev-Zel'dovich effect in

12: %the hot electron gas of galaxy clusters}

13: \author{Jian Hu\altaffilmark{1} and Yu-Qing Lou\altaffilmark{1,2,3}}

14: \altaffiltext{1}{Physics Department and the Tsinghua Center for

15: Astrophysics (THCA), Tsinghua University, Beijing 100084, China. }

16: \altaffiltext{2}{Department of Astronomy and Astrophysics,

17: The University of Chicago, 5640 South Ellis Avenue, Chicago,

18: IL 60637, USA.}

19: \altaffiltext{3}{National Astronomical Observatories,

20: Chinese Academy of Sciences, A20, Datun Road, Beijing

21: 100012, China. }

22: 

23: %December 22, 2003 (Monday) preparation for an ApJ Letter.

24: %December 24, 2003 (Wednesday) THCA

25: %December 25, 2003 (Thursday) THCA

26: %new manuscript submission: number 18078

27: %February 27, 2004 (Friday) THCA

28: 

29: \begin{abstract}

30: This Letter explores influences of intracluster magnetic fields

31: ($\gsim 1\mu$G) submerged in the hot electron gas on classic

32: Sunyaev-Zel'dovich effect (SZE) and thermal bremsstrahlung in

33: X-ray emissions. As the Larmor frequency is much higher than all

34: collision frequencies, the presence of magnetic field may lead

35: to an anisotropic velocity distribution of hot electrons. For the

36: two-temperature relativistic Maxwell-Boltzmann distribution, we

37: compute modifications to the classical thermal SZE. Intracluster

38: magnetic fields tend to enhance the SZE with steeper radial

39: variations, which bear important consequences for cluster-based

40: estimates of cosmological parameters. By applying the magnetic

41: SZE theory to spectral observations of SZ and Chandra X-ray

42: emissions from the galaxy cluster Abell 2163, a $\sim 30-40\mu$G

43: central core magnetic field $B_0$ is predicted. For the SZ and

44: Chandra X-ray spectral observations of the Coma cluster, our

45: theoretical analysis is also consistent with an observationally

46: inferred $B_0\lsim 10\mu$G. As the magnetic SZE is redshift $z$

47: independent, this mechanism might offer a potentially important

48: and unique way of probing intracluster magnetic fields in the

49: expanding universe.

50: \end{abstract}

51: 

52: \keywords{cosmic microwave background --- cosmology: theory ---

53: galaxies: clusters: general --- magnetic fields --- plasmas ---

54: radiation mechanisms: general }

55: 

56: \section{Introduction}

57: 

58: The Sunyaev-Zel'dovich effect (SZE) in galaxy clusters offers a

59: unique and powerful observational tool for cosmological studies.

60: There has been persistent progress in detecting and imaging the

61: SZE in clusters. In view of this rapid development in SZE

62: observations, several important physical effects associated with

63: the classical thermal and kinetic SZEs (Sunyaev \& Zel'dovich

64: 1969, 1980) have been further explored for their diagnostic

65: potentials, such as relativistic effects (Rephaeli 1995), the

66: shape and finite extension of a galaxy cluster with a polytropic

67: temperature (Puy et al. 2000), halo rotation SZE (Cooray \& Chen

68: 2002; Chluba \& Mannheim 2002), Brillouin scattering

69: (Sandoval-Villalbazo \& Maartens 2001), early galactic winds

70: (Majumdar et al. 2001) and cooling flows (Schlickeiser 1991;

71: Majumdar et al. 2001; Koch et al. 2002).

72: 

73: Intracluster magnetic fields have been measured using a variety of

74: techniques and diagnostics, including synchrotron relics and halo

75: radio sources within clusters, inverse Compton X-ray emissions from

76: clusters, Faraday rotation measures of polarized radio sources either

77: within or behind clusters, and cluster cold fronts in X-ray images

78: (Clarke et al. 2001; see Carrili \& Taylor 2002 for a recent review).

79: These observations reveal that most cluster atmospheres are

80: substantially magnetized with typical field strengths of $\gsim 1\mu$G

81: and with high areal filling factors out to Mpc radii. In the cores of

82: `cooling flow' clusters (Eilek \& Owen 2002; Taylor et al. 2001) and at

83: cold fronts of merging clusters (Vikhlinin et al. 2001), magnetic fields

84: may gain intensities of $\sim 10-40\mu$G and thus become dynamically

85: important.

86: 

87: Magnetic fields in the intracluster gas allows for particle

88: acceleration processes which modify specifics of heating processes,

89: such that the electron energy distribution differs from the

90: Maxwell-Boltzmann distribution. Such stochastic acceleration

91: processes include shocks and magnetohydrodynamic (MHD) waves, etc.

92: The bremsstrahlung from a modified Maxwell-Boltzmann electron gas

93: might account for the observed X-ray spectra up to highest energies

94: of current X-ray observations (Ensslin et al. 1999; Blasi 2000a).

95: If energy injections by MHD waves are turned off, a galaxy cluster

96: gradually thermalizes with electrons approaching a Maxwell-Boltzmann

97: distribution on a rough timescale of $\sim 10^7-10^8$ yrs. As all

98: collision frequencies (Nicholson 1983) are much lower than the

99: electron Larmor frequency for the magnetized intracluster gas

100: (Sarazin 1988), the electron velocity distribution is likely

101: to be anisotropic as long as the parallel (relative to magnetic

102: field ${\bf B}$) pressure is not too much higher than the

103: perpendicular pressure (Parker 1958; Hasegawa 1975).

104: 

105: We presume the result of a partial electron thermalization

106: is a two-temperature relativistic Maxwell-Boltzmann

107: distribution, i.e. an anisotropic velocity distribution.

108: %We presume that electrons thermalize their parallel and

109: %perpendicular kinetic energies separately with a resulting

110: %two-temperature relativistic Maxwell-Boltzmann distribution.

111: This two-temperature does not mean an electron gas having two

112: components with different temperatures, but refers to the same

113: population with different average kinetic energies along and

114: perpendicular to the local magnetic field. The main thrust

115: of this Letter is to advance magnetic SZE theory in contexts of

116: Chandra X-ray and radio SZE spectral observations for galaxy

117: clusters.

118: 

119: In Section 2, we calculate the X-ray emission and SZE spectra

120: using the two-temperature relativistic Maxwell-Boltzmann

121: distribution for electron velocity. Based on both Chandra X-ray

122: and SZ spectral observations, we offer a specific prediction

123: for the galaxy cluster A2163. Finally, we discuss cosmological

124: implications of our magnetic SZE theory in Section 3.

125: 

126: %\vfill\eject

127: \section{Magnetic Sunyaev-Zel'dovich Effect }

128: 

129: Both X-ray emission and SZE spectra are sensitive to the hot

130: electron energy distribution. By the presence of $\mathbf B\gsim

131: 1\mu$G, electrons thermalize their parallel and perpendicular

132: (relative to ${\bf B}$) kinetic energies separately with a

133: resulting two-temperature Maxwell-Boltzmann distribution on

134: timescales of $\sim 10^7-10^8$ yrs. We adopt a two-temperature

135: thermal relativistic Maxwellian electron velocity distribution

136: $p_{\rm e}(\beta_1,\beta_2)$, namely

137: \begin{equation}

138: p_{\rm e}(\beta_1,\beta_2)d\beta_1{\rm d}\beta_2\propto

139: \gamma^5\beta_2\exp\bigg(-\frac{\gamma_1}{\Theta_1}

140: -\frac{\gamma_2}{\Theta_2}\bigg){\rm d}\beta_1{\rm d}\beta_2\ ,

141: \end{equation}

142: where $\Theta_1\equiv k_{\rm B}T_{\parallel}/(m_{\rm e}c^2)$,

143: $\Theta_2\equiv k_{\rm B}T_{\perp}/(m_{\rm e}c^2)$, $k_{\rm B}

144: =1.38\times 10^{-16}\hbox{ erg K}^{-1}$ is the Boltzmann constant,

145: $c$ is the speed of light and $m_{\rm e}$ is the electron mass;

146: $T_{\parallel}$ and $T_{\perp}$ are parallel and perpendicular

147: temperatures, respectively; $\beta_1\equiv v_{\parallel}/c$,

148: $\beta_2\equiv v_{\perp}/c$, $\beta^2=\beta_1^2+\beta_2^2$,

149: $\gamma=(1-\beta^2)^{-1/2}$ and $\gamma_i=(1-\beta_i^2)^{-1/2}$

150: for $i=1,2$ with $v_{\parallel}$ and $v_{\perp}$ being parallel

151: and perpendicular velocities, respectively. We assume $\mathbf B$

152: to be random over the entire cluster on scales larger than a

153: typical coherence length of $\sim 1-10$kpc. Thus microscopically

154: anisotropic electrons are macroscopically isotropic, analogous

155: to a demagnetized ferromagnet. Integrating in all directions,

156: the electron speed or energy distribution becomes

157: \begin{equation}

158: p_{\rm e}(\beta){\rm d}\beta=N\gamma^5\beta {\rm d}\beta

159: \int_{-\beta}^{\beta}\exp\bigg(-\frac{\gamma_1}{\Theta_1}

160: -\frac{\gamma_2}{\Theta_2}\bigg){\rm d}\beta_1\ ,

161: \end{equation}

162: where $N$ is a normalization factor computed numerically.

163: 

164: The firehose stability (Parker 1958; Hasegawa 1975) for

165: velocity anisotropy requires

166: \begin{equation}

167: B^2/(2\pi)\gsim n_{\rm e}(\langle mv_\parallel^2\rangle

168: -\langle mv_\perp^2\rangle)>0\ ,

169: \end{equation}

170: where $n_{\rm e}$ is the electron number density,

171: $m\equiv\gamma m_{\rm e}$, angled brackets indicate averages

172: over $p_{\rm e}(\beta)$ and the mean temperature $T$ is

173: defined by $ 3k_{\rm B}T/2\equiv (2\langle mv_\perp^2\rangle

174: +\langle mv_\parallel^2\rangle)/2$. It then follows that

175: \begin{eqnarray}

176:  k_{\rm B}T_{\parallel}\equiv\langle mv_{\parallel}^2\rangle

177: =k_{\rm B}T+B^2/(3\pi n_{\rm e})\nonumber\ ,\\

178:  k_{\rm B}T_{\perp}\equiv\langle

179: mv_{\perp}^2\rangle =k_{\rm B}T-B^2/(6\pi n_{\rm e})\nonumber

180: \end{eqnarray}

181: with $B^2/(2\pi n_{\rm e}k_{\rm B})$ being the upper bound

182: for the temperature difference $T_\parallel-T_\perp$. For an

183: observed X-ray energy spectrum and an empirically inferred $B$

184: distribution in a cluster, one may estimate $T_\parallel$ and

185: $T_\perp$ by fitting the spectral data. By correlations between

186: X-ray surface brightness and Faraday rotation measure (Dolag et

187: al. 2001), a power law $B\propto [n_{\rm e}(r)]^{\alpha}$ was

188: inferred with an exponent $\alpha$ estimated from the slope of

189: $\ln B$ versus $\ln n_{\rm e}$ relation. For example, one finds

190: $\alpha\sim 0.9$ for the galaxy cluster A119 and $\alpha\sim 0.5$

191: for the galaxy cluster 3C 129 with a larger uncertainty in the

192: latter. For $\alpha=0.5$, $T_\parallel$ and $T_\perp$ can remain

193: constant in a magnetized galaxy cluster.

194: 

195: The X-ray emission rate per unit volume

196: per unit energy interval is given by

197: \begin{equation}

198: j_{\rm X}(E_{\rm X})=n_{\rm e}^2 c\int {\rm d}\beta p_{\rm

199: e}(\beta)\beta\sigma_{\rm B}(\beta ,E_{\rm X}),

200: \end{equation}

201: where $\sigma_{\rm B}(\beta,E_X)$ is the differential cross

202: section (Haug 1997) for the bremsstrahlung of an X-ray photon with

203: energy $E_{\rm X}$ from an electron of speed $c\beta$. Given a

204: central $B_0$, we fit an observed X-ray energy spectrum with the

205: electron distribution (2) and $T_\parallel-T_\perp$ constrained

206: by marginal firehose stabilities (Parker 1958; Hasagawa 1975).

207: 

208: To scatter a cosmic microwave background (CMB) photon (Birkinshaw

209: 1999) of frequency $\nu_i$ off an isotropic distribution of

210: thermal electrons with speed $c\beta$, the probability for the

211: scattered photon of frequency $\nu_i(1+s)$ is

212: \begin{eqnarray}

213: P(s,\beta)=\frac{3}{16\gamma^4\beta}\int^{\mu_2}_{\mu_1}

214: \frac{(1+\beta\mu')}{(1-\beta\mu)^{3}}

215: \qquad\qquad\qquad\qquad\nonumber\\

216: \times[1+\mu^2\mu'^2+(1-\mu^2)(1-\mu'^2)/2]{\rm d}\mu\ ,

217: \quad

218: \end{eqnarray}

219: with $\mu'\equiv [(1+s)(1-\beta\mu)-1]/\beta$, where

220: $\mu_1=(s-\beta)/[(1+s)\beta]$ and $\mu_2=1$ for $s\geq 0$ and

221: $\mu_1=-1$ and $\mu_2=(s+\beta)/[(1+s)\beta]$ for $s<0$. As the

222: intracluster electron gas is macroscopically isotropic and has a

223: thin optical depth $\tau_{\rm e}\sim 10^{-2}$, integral (5) is

224: applicable to the magnetic SZE analysis. For photons scattered by

225: an electron distribution of expression (2), the resulting

226: distribution in the fractional frequency shift $s$ is

227: \begin{equation}

228: P_1(s)=\int^1_{s/(2+s)}{\rm d}\beta p_{\rm e}(\beta)P(s,\beta)\ .

229: \end{equation}

230: For CMB photons scattered by a hot intracluster

231: electron gas, the change in the CMB spectrum at

232: frequency $\nu$ caused by the magnetic SZE is

233: \begin{equation}

234: \Delta I(\nu)={2h\nu^3\over c^2}\tau_{\rm e}

235: \int^{+\infty}_{-1}\!\!\!\!\!{\rm d}s\bigg[{P_1(s)(1+s)^3

236: \over e^{(1+s)x}-1}-{P_1(s)\over e^x-1}\bigg],

237: \end{equation}

238: where $x\equiv h\nu/(k_{\rm B}T_{\rm CMB})$, $h=6.63\times

239: 10^{-27}\hbox{erg s}$ is the Planck constant, $T_{\rm CMB}$ is the

240: present CMB temperature and $\tau_{\rm e}=\sigma_{\rm T}N_{\rm e}$

241: with $\sigma_{\rm T}=6.65\times 10^{-25}\hbox{ cm}^2$ being the

242: electron Thomson cross section (Rybicki \& Lightman 1979)

243: and $N_{\rm e}$ being the column density of free electrons

244: along the line of sight.

245: 

246: In the extensively used beta-model for intracluster hot

247: electron gas (Sarazin 1988; Fabian 1994), the empirical

248: radial distribution of $n_{\rm e}(r)$ is

249: \begin{equation}

250: n_{\rm e}(r)=n_{\rm e0}\big[1 +(r/r_{\rm c})^2\big]^{-3\beta_{\rm

251: c}/2}\ ,

252: \end{equation}

253: where $r_{\rm c}$ is the core radius and $n_{\rm e0}$ is the

254: central electron number density. It follows that $\tau_{\rm e}$

255: at radius $r$ is given by

256: \begin{equation}

257: \tau_{\rm e}(r)=\tau_{\rm e0}\big[1 +(r/r_{\rm

258: c})^2\big]^{(1-3\beta_{\rm c})/2}\ ,

259: \end{equation}

260: where $\tau_{\rm e0}=n_{\rm e0}\sigma_{\rm T}r_{\rm c}

261: \sqrt{\pi}\Gamma(3\beta_{\rm c}-1/2)/\Gamma(3\beta_{\rm c})$ and

262: $\Gamma$ is the gamma function.\footnote{Here, $\beta_{\rm c}$

263: instead of $\beta$ is used to avoid notation confusions with the

264: dimensionless thermal electron speed $\beta$.} We use eqns. (2),

265: (5), (6) and (9) in integral (7) to compute magnetic SZE

266: spectrum at $r$.

267: 

268: \begin{figure}

269: \includegraphics[angle=0,totalheight=4.5cm,width=7.0cm,scale=0.4]

270: %{a2163_sze.eps} \twocolumn \caption{The SZE spectrum of A2163

271: {a2163_sze.eps} \caption{The SZE spectrum of A2163 (triangles with

272: error

273: %\vskip -0.1cm

274: bars are data points). The heavy solid line ($T=12.4$ keV) is the

275: thermal SZE with parameters determined from X-ray observations with

276: $B_{0}=0$. The dash, dotted and dash-dotted lines represent magnetic

277: SZE with different central magnetic field strengths $B_0$ and with

278: other cluster parameters being the same. We take $\alpha=0.5$, so

279: that $T_{\perp}$ and $T_{\parallel}$ remain constant in the

280: cluster. The grey line ($T=14.4$keV, $B_{0}=0$) coincides with

281: the dash-dotted line.}

282: %\vskip 0.7cm

283: \end{figure}

284: 

285: \begin{figure}

286: \includegraphics[angle=0,totalheight=5.0cm,width=8.0cm,scale=0.9]

287: {a2163_spe.eps} \caption{Energy spectrum of the central $r\lsim5'$

288: region of A2163 by Chandra. The grey and dashed lines are the

289: $T=12.4$ and $14.4$ keV thermal bremsstrahlung fits. The dotted

290: line shows our magnetic model of essentially the same fit as

291: $T=12.4$ keV yet with different $B_0$, $T_{\parallel}$ and

292: $T_{\perp}$ of Fig. 1. The lower energy part ($\lsim 2.1$ keV) is

293: contaminated by the soft excess (Markevitch \& Vikhlinin 2001).

294: The spectral resolution for energy $\gsim 5.0$ keV is not high

295: enough and is blended with the Fe K$_\alpha$ line (peaked at 5.3

296: keV for $z=0.203$). We fit the $2.1-5.0$ keV band with

297: uncertainity in $T$ of 0.9 keV (90\%CL). }

298: %\vskip 0.7cm

299: \end{figure}

300: 

301: \begin{figure}

302: \includegraphics[angle=0,totalheight=4.5cm,width=8.0cm,scale=0.4]

303: %{a2163_szed.eps} \twocolumn \caption{Radial features of thermal

304: {a2163_szed.eps} \caption{Radial features of thermal and magnetic

305: SZEs modeled for A2163 at $\nu_0=30$GHz. The solid line is the

306: thermal SZE, the dash, dotted and dash-dotted lines are models

307: with the same central magnetic field strength $B_0=36\mu $G but

308: different $\alpha$ values. }

309: \end{figure}

310: 

311: For spectral observations of A2163, we estimate the lower limit of

312: intracluster $B$ from data using our magnetic SZE theory. Being

313: one of the hottest clusters, the thermal electron gas trapped in

314: A2163 (redshift $z=0.203$) has a mean temperature $k_{\rm B}T_{\rm

315: c}\simeq 12.4\pm 0.5$keV and a central density $n_{\rm e0}\simeq

316: 6.82\times 10^{-3} {\rm cm}^{-3}$ with a core radius $r_{\rm c}=

317: 0.269\pm 0.025$Mpc (Hubble constant $h_{100}=0.71$) and a

318: $\beta_{\rm c}=0.616\pm 0.031$ (Elbaz et al. 1995; Markevitch et

319: al. 1996; Markevitch \& Vikhlinin 2001). By these estimates, the

320: optical depth towards the cluster center is $\tau_{\rm e0}=0.0133$

321: and the central Compton parameter is $y_{\rm th0}=3.21\times 10^{-4}$.

322: 

323: A conspicuous SZE spectrum has been observed in A2163 at four

324: frequencies by BIMA at 30GHz (LaRoque et al. 2002), by DIABOLO

325: at 140GHz (Desert et al. 1998) and by SuZIE at 140, 218 and

326: 270GHz (Holzapfel et al. 1997) with dust-corrections (LaRoque

327: et al. 2002). These data were previously fit with a Compton

328: parameter $y_{\rm th}=(3.56^{+0.41+0.27}_{-0.41-0.19})\times

329: 10^{-4}$ for a thermal SZE together with a kinetic SZE for a

330: positive peculiar velocity

331: $V_{\rm p}=415_{-850\ -440}^{+1030+460}\ {\rm km\ s^{-1}}$

332: (68\%CL) (Carlstrom et al. 2002). A positive-velocity kinetic

333: SZE leads to an overall downward shift of the thermal SZE

334: spectrum especially around the zero point ($\sim218$ GHz)

335: of the SZE.

336: 

337: We fit the observed X-ray spectrum of Fig. 2 for a mean

338: temperature $T\cong 12.4$keV by eqns. (2) and (4) with different

339: $B$ strengths and infer relations among $B,\  T_\parallel

340: \hbox{ and }T_\perp$. We here take $\alpha=0.5$ with $T_{\perp}$

341: and $T_{\parallel}$ being constants (see parameters of Fig. 1).

342: Inserting these relations into eqns. (2), (6) and (7), we obtain

343: the spectra of magnetic SZE.

344: 

345: Shown in Fig. 1 are SZE spectra computed for A2163 with different

346: parameters in comparison with the observed SZE spectrum. The model

347: without magnetic field (solid line for $T=12.4$keV) underestimates

348: signals of SZE (especially at $\nu=140$GHz) with $\chi^2=3.82$

349: in a chi-square fit. Intracluster magnetic field tends to enhance

350: signals of SZE and the best fit is $B_0=36\mu$G with $\chi^2=0.78$.

351: As the marginal firehose stability (Parker 1958) is used here, the

352: fitting estimate represents the lower limit of $B_0=36\mu$G for

353: A2163. Magnetic field also increases the null frequency of the

354: pure thermal SZE, similar to the kinetic SZE with a positive

355: velocity. This degeneracy can be removed by SZE signals at other

356: frequencies (e.g. 100 and 400GHz, etc.). The SZE spectrum data

357: can be fit with $T=14.4$keV (grey line in Fig. 1) and $B_0=0$,

358: but the thermal spectrum data does not fit well with $T=14.4$keV

359: (dashed line in Fig. 2). Shown in Fig. 3 are computed radial SZE

360: features of thermal ($B=0$) and magnetic ($\alpha\neq 0$) SZEs

361: modeled for cluster A2163 at 30GHz. With a smaller exponent $\alpha$,

362: SZE signals steepen from the central to peripheral parts of a cluster.

363: This can be utilized to determine macroscopic mean $\mathbf B$

364: structures by obtaining spatially resolved magnetic SZE intensity

365: maps (Carlstrom et al. 2002). Note that 30GHz of Fig. 3 is just an

366: illustrating example.

367: 

368: \section{Discussions }

369: 

370: Contrary to recent results (Koch et al. 2003; Zhang 2003), we

371: find that the anisotropic velocity distribution of electrons

372: caused by magnetic field $B$ enhances the SZE. Our model results

373: of Figs. $1-3$ can be critically tested against more precise

374: spectral SZE measurements of A2163 in the frequency bands of

375: $\sim 50-130$GHz and $\sim 300-600$GHz by MAX,

376: %$^{66}$,

377: MSAM

378: %$^{67}$

379: and SuZIE

380: %$^{68}$

381: types of experiments in the frequency passbands of $90-670$GHz and

382: by AMiBA in the band $84-104$GHz, Nobeyama at 21 and 43GHz, JCMT

383: at 350 and 650GHz, SZA

384: %The Sunyaev-Zel'dovich Array (SZA) consists of eight 3.5 m dishes and

385: %a digital correlator with 8 GHz bandwidth. It will have low-noise HEMT

386: %(high electron mobility transistor)

387: %receivers operating at 26 - 36 GHz and also at 85 - 115 GHz.

388: %James Clark Maxwell Telescope (JCMT) Submillimeter Common-User

389: %Bolometer Array (SCUBA) on JCMT.

390: in the bands of $26-36$GHz and $85-115$GHz, BIMA and OVRO in the

391: band of $26-36$GHz,

392: %Berkeley-Illinois-Maryland Association;

393: %Owens Valley Radio Observatory

394: MINT

395: %Millimeter Interferometer, observed the CMB from the Chilean

396: %Atacama Desert in late 2001: operating around 150 GHz

397: at 150GHz and ACT

398: %Atacama Cosmology Telescope (ACT): 150, 220, and 270 GHz

399: at 150, 220, and 270 GHz. Multi-frequency projects such as the

400: upgraded MITO (Lamagna et al. 2002; De Petris et al. 2002) and

401: the OLIMPO (Masi et al. 2003) experiments are very promising to

402: provide some results.

403: 

404: A2163 may involve merger shocks that could amplify $B$ (e.g.

405: Markevitch \& Vikhlinin 2001). The spectral index maps of A2163

406: show a spectral steepening from the central to peripheral radio

407: halo regions, implying a radial decrease of $B$ in reacceleration

408: models (e.g. Feretti et al. 2003). It was attempted to fit the

409: SZE spectrum of A2163 with a combination of thermal and non-thermal

410: electrons (e.g. Colafrancesco et al. 2003), but no evidence was

411: found for hard X-ray excess due to the non-thermal component in

412: the BeppoSAX data (e.g. Feretti et al. 2001).

413: 

414: Based on X-ray and SZE measurements, 41 galaxy clusters were used

415: to independently estimate Hubble constant $h_{100}=0.61\pm 0.03\pm

416: 0.18$, where the uncertainties are statistical and systematic at

417: 68\% confidence level for $\Omega_M=0.3$ and

418: $\Omega_{\Lambda}=0.7$ cosmology (Carlstrom et al. 2002; Reese 2003).

419: Our analysis of A2163 shows that intracluster magnetic field induces

420: microscopic anisotropies in electron velocity distribution to enhance

421: the SZE. It appears that inferences from cluster models without magnetic

422: field would systematically

423: underestimate $h_0$ as in the case of A2163 for which Holzapfel et

424: al. (1997) inferred a lower $h_{100}=0.60\pm 0.04$ against the

425: current WMAP result of $h_{100}=0.71\pm 0.04$. As the cluster

426: asphericity and orientation in the sky are random and the average

427: cluster peculiar velocity is zero, these factors should contribute

428: to the systematic uncertainty with the Hubble constant being

429: statistically unaltered. The underestimation of Hubble constant

430: may be explained by the generic presence of core magnetic field

431: $B_0\sim 10-40\mu$G in this sample of galaxy clusters.

432: 

433: Another important cosmological effect of the ubiquitous enhancement

434: of magnetic SZE due to the prevalence of $\gsim 1\mu$G magnetic

435: fields in galaxy clusters would be observable in the CMB angular

436: spectrum especially at high $l\gsim 3000-4000$. This contribution

437: to CMB fluctuations may be estimated and tested by CMB experiments

438: such as ACT, Planck, SZA, etc. Details of these two cosmological

439: effects will be pursued in forthcoming papers.

440: 

441: For X-ray (Arnaud et al. 2001) and SZE (De Petris et al. 2002)

442: spectral observations of Coma cluster (Abell 1656), our magnetic

443: SZE analysis is consistent with the currently inferred $B_0\lsim

444: 10\mu$G (Carilli \& Taylor 2002). Likewise, magnetic SZE can be

445: utilized in other galaxy clusters with high-resolution and

446: high-sensitivity X-ray and SZE spectral observations to estimate

447: the lower limit of $B_0$ as well as SZE spatial features. While it

448: is necessary to estimate all possible corrections to the classic

449: SZE in order to isolate the magnetic contribution, this may be a

450: unique procedure to probe intracluster magnetic fields at high

451: redshifts $z$, at least statistically.

452: %Multi-frequency projects such as the upgraded MITO (Lamagna

453: %et al. 2002) and the OLIMPO (Masi et al. 2003) experiments

454: %may give some preliminary results.

455: Finally, anisotropic distributions of nonthermal electrons

456: should lead to distinct magnetic SZE in radio lobes of

457: extragalactic jets.

458: \\

459: 

460: We thank the referee Dr. D. Puy for useful comments. This

461: research has been supported in part by the ASCI Center for

462: Astrophysical Thermonuclear Flashes at the U. of Chicago under

463: DOE contract B341495, by the Special Funds for Major State Basic

464: Science Research Projects of China, by the THCA, by the

465: Collaborative Research Fund from the NSF of China for Outstanding

466: Young Overseas Chinese Scholars (NSFC 10028306) at the National

467: Astronomical Observatory, CAS, by NSFC grant 10373009 at the

468: Tsinghua U., and by the Yangtze Endowment from the Ministry of

469: Education through the Tsinghua U. Affiliated institutions of YQL

470: share this contribution.

471: 

472: %\onecolumn

473: \begin{thebibliography}{}

474: \bibitem{}Arnaud, M., et al. 2001, A\&A, 365, L67

475: \bibitem{}Birkinshaw, M. 1999, Phys. Rep., 310, 97

476: \bibitem{}Blasi, P. 2000a, ApJ, 532, L9

477: \bibitem{}Blasi, P. Olinto, A. V., Stebbins, A. 2000b, ApJ, 535, L71

478: \bibitem{}Carlstrom, J., Holder, G., Reese, E. 2002, \araa, 40, 643

479: \bibitem{}Carilli, C. L., Taylor, B. 2002, \araa, 40, 319

480: \bibitem{}Chluba, J., Mannheim, K. 2002, A\&A, 396, 419

481: \bibitem{}Clarke, T. E., Kronberg, P. P., B\"ohringer, H.

482: 2001, ApJ, 547, L111

483: %A new radio-X-ray probe of galaxy

484: %cluster magnetic fields. -L114 (2001).

485: \bibitem{}Colafrancesco, S., Marchegiani, P., Palladino, E.

486: 2003, A\&A, 397, 27

487: \bibitem{}Cooray, A., Chen, X. 2002, ApJ, 573, 43

488: \bibitem{}De Petris, M. et al. 2002, ApJ, 574, L119

489: \bibitem{}Desert, F. et al. 1998, New Astron., 3, 655

490: \bibitem{}Dolag, K. et al.

491: %, Schindler, S., Govoni, F. \& Feretti, L.

492: 2001, A\&A, 378, 777

493: \bibitem{}Eilek, J.A., Owen, F.N. 2002, ApJ, 567, 202

494: \bibitem{}Elbaz, D., Arnaud, M., B\"oringer, H. 1995, A\&A, 293, 337

495: \bibitem{}Fabian, A. C. 1994, ARA\&A, 32, 277

496: %Cooling flows in clusters of galaxies.

497: %{\it Annu. Rev. Astron. Astrophys.} -318 (1994).

498: \bibitem{}Feretti, L. et al.

499: %, Fusco-Femiano, R., Giovannini, G. \& Govoni, F.

500: 2001, A\&A, 373, 106

501: \bibitem{}Feretti, L. et al. 2003, The Cosmic Cauldron,

502: 25th IAU meeting

503: \bibitem{}Haug E. 1997, A\&A, 326, 417

504: \bibitem{}Hasegawa, A. 1975, Plasma Instability

505: and Nonlinear Effects, Springer-Verlag, Berlin

506: \bibitem{}Holzapfel, W. L. et al. 1997, ApJ, 480, 449

507: \bibitem{}Kaiser, N. 1986, \mnras, 222, 323

508: \bibitem{}Koch, P. M., Jetzer, Ph., Puy, D. 2002, New Astron., 7, 587

509: \bibitem{}Koch, P. M., Jetzer, Ph., Puy, D. 2003, New Astron., 8, 1

510: \bibitem{}Lamagna, L. et al. 2002, AIP Conf. Proc. 616, 92

511: \bibitem{}LaRoque, S. J. et al. 2002, ApJ, submitted (astro-ph/0204134)

512: \bibitem{}Majumdar, S., Nath B. Chiba M., 2001, MNRAS, 324, 537

513: \bibitem{}Markevitch, M. et al. 1996, ApJ, 456, 437

514: \bibitem{}Markevitch, M., Vikhlinin A. 2001, ApJ, 563, 95

515: \bibitem{}Masi, S. et al. 2003, Mem.S.A.It 74, 96

516: \bibitem{}Nicholson, D. R. 1983, Introduction

517: to Plasma Theory, Wiley, New York

518: %Dwight

519: \bibitem{}Parker, E. N. 1958, Phys. Rev., 109, 1874

520: %Dynamical instability of an anisotropic ionized gas of low density.

521: \bibitem{}Puy, D., Grenacher, L., Jetzer, Ph., Signore, M. 2000, A\&A 363, 415

522: \bibitem{}Reese, E. D. 2003, Measuring and Modeling the Universe,

523: ed. W. L. Freedman (Cambridge: Cambridge Univ. Press)

524: \bibitem{}Rephaeli, Y. 1995, ARA\&A, 33, 541

525: \bibitem{}Rosati, P., Borgani, S., Norman, C. 2002, \araa, 40, 539

526: \bibitem{}Rybicki, G. B., Lightman, A. P. 1979,

527: Radiative Processes in Astrophysics, Wiley, New York

528: \bibitem{}Sandoval-Villalbazo, A., Maartens, R. 2001, astro-ph/0105323

529: \bibitem{}Sarazin, C. 1988, X-Ray Emission from

530: Clusters of Galaxies, Camb. Univ. Press, Cambridge

531: \bibitem{}Schlickeiser, R. 1991, A\&A, 248, L23

532: \bibitem{}Sunyaev, R. A., Zel'dovich, Y. B. 1969, Ap\&SS, 4, 301

533: \bibitem{}Sunyaev, R. A., Zel'dovich, Y. B. 1980, ARA\&A, 18, 537

534: \bibitem{}Taylor, G. B., Fabian, A. C., Allen, S. W. 2002, MNRAS, 334, 769

535: \bibitem{}Vikhlinin, A., Markevitch, M., Murray, S. S. 2001, ApJ, 549, L47

536: %\bibitem{}Zeldovich, Y. B. \& Sunyaev, R. A. 1969, Ap\&SS, 4, 301

537: \bibitem{}Zhang, P. J. 2003, astro-ph/0308354

538: \end{thebibliography}

539: 

540: \end{document}

541: