1: 

2: \documentclass[11pt]{article}  

3: \usepackage{graphicx,floatflt,amssymb}    

4: \usepackage{epsfig,axodraw}    

5: \usepackage{graphics}    

6: \usepackage{psfrag}    

7: \textwidth=17cm    

8: \textheight=22.5cm     

9: \oddsidemargin -0.3cm     

10: \topmargin -1.5cm     

11: \parskip 0.3cm     

12: \tolerance=10000     

13: \parindent 0pt     

14: 

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

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

17: \newcommand{\bd}{\begin{displaymath}}     

18: \newcommand{\ed}{\end{displaymath}}     

19: \newcommand{\be}{\begin{equation}}     

20: \newcommand{\ee}{\end{equation}}     

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

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

23: 

24:      

25: \newcommand{\rpv}{\mbox{$\not \hspace{-0.10cm} R_p$ }}

26: \def\ltap{\raisebox{-.4ex}{\rlap{$\sim$}} \raisebox{.4ex}{$<$}}     

27: \def\gtap{\raisebox{-.4ex}{\rlap{$\sim$}} \raisebox{.4ex}{$>$}}     

28: 

29:      

30: 

31: \def\Rpv{R_p \! \! \! \! \! \! /~~}

32: \def\beq{\begin{equation}}

33: 

34: \def\eeq{\end{equation}}

35: 

36: \def\bea{\begin{eqnarray}}

37: 

38: \def\eea{\end{eqnarray}}

39: 

40: \def\bq{\begin{quote}}

41: 

42: \def\eq{\end{quote}}

43: 

44: \def \lsim{\mathrel{\vcenter

45: 

46:      {\hbox{$<$}\nointerlineskip\hbox{$\sim$}}}}

47: 

48: \def \gsim{\mathrel{\vcenter

49: 

50:      {\hbox{$>$}\nointerlineskip\hbox{$\sim$}}}}

51: 

52: \def\gappeq{\mathrel{\rlap {\raise.5ex\hbox{$>$}}

53: 

54: {\lower.5ex\hbox{$\sim$}}}}

55: 

56: \def\lappeq{\mathrel{\rlap{\raise.5ex\hbox{$<$}}

57: 

58: {\lower.5ex\hbox{$\sim$}}}}

59: 

60: \def\simlt{\stackrel{<}{{}_\sim}}

61: 

62: \def\simgt{\stackrel{>}{{}_\sim}}

63: 

64: \def\msnu{m_{\tilde{\nu}}}

65: 

66: \def\msnui{m_{\tilde{\nu}_i}}

67: 

68: \def\msnuj{m_{\tilde{\nu}_j}}

69: 

70: \def\msnuI{m_{\tilde{\nu}_I}}

71: 

72: \def\msnuJ{m_{\tilde{\nu}_J}}

73: 

74: \def\snu{\tilde{\nu}}

75: 

76: \def\snuvi{\langle\tilde{\nu}_i\rangle}

77: 

78: \def\Huv{\langle H_u \rangle}

79: 

80: \def\mchi{m_{\chi}}

81: 

82: \def\Rp{R_p}

83: 

84: \def\Rpv{R_p \! \! \! \! \! \! /~~}

85: 

86: \def\mnu{[m_{\nu}]_{ij}}

87: 

88: \def\LI{$\{ L_I \} ~$}

89: 

90: \def\vd{\vec{\delta}}

91: 

92: \def\hd{\hat{\delta}}

93: 

94: \def\msusy{M_S}

95: 

96: \def\ltap{\raisebox{-.4ex}{\rlap{$\sim$}} \raisebox{.4ex}{$<$}}

97: 

98: \def\gtap{\raisebox{-.4ex}{\rlap{$\sim$}} \raisebox{.4ex}{$>$}}

99: 

100: \newcommand{\Rsl}{{\not \! \!{R}}}

101: 

102: \newcommand{\sqm}{m_{\tilde{q}}}

103: 

104: \newcommand{\slm}{m_{\tilde{l}}}

105: 

106: \newcommand{\ms}{\widetilde{m}}

107: 

108: \newcommand{\onetwo}{\Delta m^2_{12}}

109: 

110: \newcommand{\onethree}{\Delta m^2_{13}}

111: 

112: \newcommand{\twothree}{\Delta m^2_{23}}

113: 

114: \newcommand{\sun}{\Delta m^2_{\rm solar}}

115: 

116: \newcommand{\atm}{\Delta m^2_{\rm atm}}

117: 

118: %\newcommand{\lsnd}{\Delta m^2_{\rm LSND}}

119: 

120: \newcommand{\ssqsun}{\sin^2 2\theta_{12}}

121: 

122: \newcommand{\ssqatm}{\sin^2 2\theta_{23}}

123: 

124: 

125: 

126: \newcommand{\dmui}{\delta^i_{\mu}}

127: 

128: \newcommand{\dmuj}{\delta^j_{\mu}}

129: 

130: \newcommand{\dbi}{\delta^i_{B}}

131: 

132: \newcommand{\dbj}{\delta^j_{B}}

133: 

134: \newcommand{\dlijk}{\delta^{ijk}_{\lambda}}

135: 

136: \newcommand{\dlpipq}{\delta^{ipq}_{\lambda'}}

137: 

138: 

139: 

140: \def\bea{\begin{eqnarray}}   

141: 

142: \def\eea{\end{eqnarray}}

143: 

144: \def\la{\mathrel{\mathchoice {\vcenter{\offinterlineskip\halign{\hfil

145: 

146: $\displaystyle##$\hfil\cr<\cr\sim\cr}}}

147: 

148: {\vcenter{\offinterlineskip\halign{\hfil$\textstyle##

149: 

150: $\hfil\cr<\cr\sim\cr}}}

151: 

152: {\vcenter{\offinterlineskip\halign{\hfil$\scriptstyle##

153: 

154: $\hfil\cr<\cr\sim\cr}}}

155: 

156: {\vcenter{\offinterlineskip\halign{\hfil$\scriptscriptstyle##$

157: 

158: \hfil\cr<\cr\sim\cr}}}}}

159: 

160: % greater than or order of \ga

161: 

162: \def\ga{\mathrel{\mathchoice {\vcenter{\offinterlineskip\halign{\hfil

163: 

164: $\displaystyle##$\hfil\cr>\cr\sim\cr}}}

165: 

166: {\vcenter{\offinterlineskip\halign{\hfil$\textstyle##

167: 

168: $\hfil\cr>\cr\sim\cr}}}

169: 

170: {\vcenter{\offinterlineskip\halign{\hfil$\scriptstyle##

171: 

172: $\hfil\cr>\cr\sim\cr}}}

173: 

174: {\vcenter{\offinterlineskip\halign{\hfil$\scriptscriptstyle##$

175: 

176: \hfil\cr>\cr\sim\cr}}}}}

177: 

178:      

179: \def\Journal#1#2#3#4{{#1} {\bf #2}, #3 (#4)} 

180:  

181: % Some useful journal names 

182: \def\NCA{\em Nuovo Cimento} 

183: \def\NIM{\em Nucl. Instrum. Methods} 

184: \def\NIMA{{\em Nucl. Instrum. Methods} A} 

185: \def\NPB{{\em Nucl. Phys.} B} 

186: \def\PLB{{\em Phys. Lett.}  B} 

187: \def\PRL{\em Phys. Rev. Lett.} 

188: \def\PRD{{\em Phys. Rev.} D} 

189: \def\ZPC{{\em Z. Phys.} C} 

190: \def\PR{\em Phys. Rep.}  

191: \def\JCAP{\em JCAP}

192: \def\JHEP{\em JHEP}  

193: 

194: %\newcommand{\sun}{\Delta m^2_{\rm sol}}     

195: %\newcommand{\atm}{\Delta m^2_{\rm atm}}     

196: %\newcommand{\lsnd}{\Delta m^2_{\rm LSND}}     

197: %\newcommand{\nonu}{(0\nu\beta\beta)}     

198: %\newcommand{\meff}{|m_{ee}|}     

199: %\newcommand{\angsun}{\sin^2 2\theta_{\rm sol}}     

200: %\newcommand{\angchz}{\sin \theta_{13}}     

201: %\newcommand{\l}{\el}     

202: 

203:      

204: 

205: %less than or order of     

206: \def\la{\mathrel{\mathchoice {\vcenter{\offinterlineskip\halign{\hfil     

207: $\displaystyle##$\hfil\cr<\cr\sim\cr}}}     

208: {\vcenter{\offinterlineskip\halign{\hfil$\textstyle##     

209: $\hfil\cr<\cr\sim\cr}}}     

210: {\vcenter{\offinterlineskip\halign{\hfil$\scriptstyle##     

211: $\hfil\cr<\cr\sim\cr}}}     

212: {\vcenter{\offinterlineskip\halign{\hfil$\scriptscriptstyle##$     

213: \hfil\cr<\cr\sim\cr}}}}}     

214: % greater than or order of \ga     

215: \def\ga{\mathrel{\mathchoice {\vcenter{\offinterlineskip\halign{\hfil     

216: $\displaystyle##$\hfil\cr>\cr\sim\cr}}}     

217: {\vcenter{\offinterlineskip\halign{\hfil$\textstyle##     

218: $\hfil\cr>\cr\sim\cr}}}     

219: {\vcenter{\offinterlineskip\halign{\hfil$\scriptstyle##     

220: $\hfil\cr>\cr\sim\cr}}}     

221: {\vcenter{\offinterlineskip\halign{\hfil$\scriptscriptstyle##$     

222: \hfil\cr>\cr\sim\cr}}}}}     

223: 

224:      

225: 

226:      

227: 

228: \begin{document}     

229: \vspace*{-0.5in}     

230: \renewcommand{\thefootnote}{\fnsymbol{footnote}}     

231: \begin{flushright}     

232: LPT Orsay/06-27 \\     

233: %\texttt{hep-ph/yymmddd}      

234: \end{flushright}     

235: \vskip 5pt     

236: 

237: 

238: \begin{center}     

239: {\Large {\bf An origin for small neutrino masses in the NMSSM}}

240: \vskip 25pt     

241: {\bf Asmaa Abada $^{1,}$\footnote{E-mail

242:  address: abada@th.u-psud.fr}}, 

243: {\bf Gr\'egory Moreau $^{2,}$\footnote{E-mail

244:  address: greg@cftp.ist.utl.pt}}

245: \vskip 10pt

246: $^1${\it Laboratoire de Physique Th\'eorique, Universit\'e

247: de Paris-sud XI} \\ 

248: {\it B\^atiment 210, 91405 Orsay, France} \\ 

249: $^2${\it CFTP, Departamento de F\'{\i}sica,

250: Instituto Superior T\'ecnico} \\ 

251: {\it Av. Rovisco Pais 1,

252: 1049-001 Lisboa, Portugal} \\ 

253: 

254: \normalsize     

255: \end{center}

256: 

257: 

258: \begin{abstract}  

259:  

260: We consider the Next to Minimal Supersymmetric Standard Model (NMSSM) which provides

261: a natural solution to the so-called $\mu$ problem by introducing a new gauge-singlet 

262: superfield $S$.  

263: We realize that a new mechanism 

264: of neutrino mass suppression, based on the R-parity violating bilinear terms $\mu_i L_i H_u$ 

265: mixing neutrinos and higgsinos, arises within the NMSSM, 

266: offering thus an original solution to the neutrino mass problem (connected to the 

267: solution for the $\mu$ problem). We 

268: generate realistic (Majorana) neutrino mass values without requiring any strong hierarchy amongst 

269: the fundamental parameters, in contrast with the alternative models. In particular, the 

270: ratio $| \mu_i / \mu |$ can reach $\sim 10^{-1}$, unlike in the MSSM where it has to be 

271: much smaller than unity. We check that the obtained parameters also satisfy the collider 

272: constraints and internal consistencies of the NMSSM. The price to pay for this new 

273: cancellation-type mechanism of neutrino mass reduction is a certain fine tuning, which 

274: get significantly improved in some regions of parameter space. Besides, we discuss the 

275: feasibility of our scenario when the R-parity violating bilinear terms have a common origin 

276: with the $\mu$ term, namely when those are generated via a VEV of the $S$ scalar component 

277: from the couplings $\lambda_i S L_i H_u$. Finally, we make comments on some specific 

278: phenomenology of the NMSSM in the presence of R-parity violating bilinear terms.

279: \\

280: \vskip 5pt \noindent  

281: \texttt{PACS Ns: 11.30.Fs, 12.60.Jv, 14.60.Pq, 14.80.Ly} \\  

282: \texttt{Keywords: NMSSM, Neutrino Physics, R-parity violation}  

283: 

284: \end{abstract}  

285: 

286: 

287: \newpage 

288: 

289: 

290: \renewcommand{\thesection}{\Roman{section}}  

291: \setcounter{footnote}{0}  

292: \renewcommand{\thefootnote}{\arabic{footnote}}  

293: 

294: 

295: 

296: 

297: \section{Introduction}

298: \label{intro}

299: 

300: 

301: The most severe theoretical drawback of the Standard Model (SM) is probably 

302: the gauge hierarchy problem (see for example \cite{Susskind}). In well defined supersymmetric extensions 

303: of the SM, the property of cancellation of quadratic divergences allows to address this problem. With regard

304: to the field content, the most economical candidate for such a realistic extension is the Minimal Supersymmetric

305: Standard Model (MSSM). Nevertheless, within the MSSM, there are two unexplained hierarchies. 

306: \\ The first one is intrinsic to supersymmetric models: it is named as the $\mu$ problem \cite{mu}. It arises 

307: from the presence of a mass ($\mu$) term for the Higgs fields in the superpotential. The only two natural values 

308: for this $\mu$ parameter are either zero or the Planck energy scale. While the former value is excluded by experiments

309: as it gives rise to the unacceptable existence of an axion, the latter one reintroduces the gauge hierarchy problem.

310: \\ The other hierarchy with an unknown origin is the one existing between the small neutrino masses and 

311: the electroweak symmetry breaking scale ($\sim 100 \mbox{GeV}$). Indeed, during last years, neutrino oscillation

312: experiments have confirmed that neutrinos are massive. Furthermore, the additional results, extracted from 

313: tritium beta decay experiments and cosmological data, indicate that the values of absolute neutrino 

314: masses are typically smaller than the $\mbox{eV}$ scale.

315: 

316: In this paper, 

317: we propose a supersymmetric scenario which has the virtue of addressing simultaneously 

318: both of these hierarchy questions: the $\mu$ value naturalness {\it and} the neutrino mass 

319: smallness. A nice feature of our scenario is that the mechanisms explaining the two hierarchy origins 

320: are connected, since they involve the same additional gauge-singlet superfield, providing thus

321: a common source to the solutions of these two independent problems. 

322: \\ Our framework is the Next to Minimal Supersymmetric Standard Model (NMSSM) \cite{NMSSM}

323: \footnote{The phenomenology of the NMSSM was studied, for instance, in \cite{NMpheno}.}. 

324: The NMSSM provides an 

325: elegant solution to the $\mu$ problem through the introduction of a new gauge-singlet superfield $S$ 

326: entering the scale invariant superpotential. The scalar component of $S$ acquires naturally a Vacuum 

327: Expectation Value (VEV) of the order of the supersymmetry breaking scale, generating an effective $\mu$ 

328: parameter of order of the electroweak scale. Another appealing feature of the NMSSM is to soften the

329: ``little fine tuning problem'' of the MSSM \cite{littleFT}. The introduction of suitable

330: non-renormalizable operators \cite{operators} can avoid the possibility of a cosmological domain wall 

331: problem \cite{wall}. There exist different explanations for a $\mu$ value of order of the electroweak

332: scale, but those arise in extended frameworks.

333: \\ In supersymmetric extensions of the SM, there exist coupling terms violating the so-called R-parity 

334: symmetry \cite{rpar1,rpar2}  

335: which acts on fields like $(-1)^{3B+L+2S}$, $B$, $L$ and $S$ being respectively

336: the Baryon number, Lepton number and Spin. From a purely theoretical point of view, these terms must 

337: be considered, even if some phenomenological limits apply on the R-parity violating (\rpv)

338: coupling constants \cite{reviewsD,reviewsB,HalfWay,Marc,PhysRep}. 

339: As a matter of fact, these terms are supersymmetric, gauge invariant

340: and some of them are renormalizable.  

341: Moreover, from the points of view of scenarios with discrete gauge symmetries \cite{Ibanez}, 

342: Grand Unified Theories (GUT) \cite{AlignI}-\cite{GUT1} as well as string theories \cite{string}, 

343: there exists no fundamental argument against the violation of the R-parity symmetry \cite{reviewsD}. 

344: In the present work, we consider the `bilinear' R-parity violating term

345: $H_u L$ appearing in the superpotential, $H_u$ and $L$ being respectively the up Higgs and

346: lepton doublet superfields. The existence/influence of the other \rpv terms will also be discussed. 

347: This bilinear interaction has been recently considered within the NMSSM context \cite{Chemtob}.

348: In particular, this type of interaction, which breaks the lepton number, mixes the higgsino and 

349: neutrinos together so that the neutrino field picks up a Majorana mass \cite{MixNeut} (the generation 

350: of such a neutrino mass requires two units of $L$ violation). Hence, no

351: additional right handed neutrino has to be introduced in order to generate a non-vanishing neutrino 

352: mass term.

353: \\ In our scenario, the smallness of absolute neutrino mass scale, with respect to electroweak scale, 

354: finds an origin in the following sense: the neutrino field acquires a mass of the $\mbox{eV}$ order  

355: without requiring any high hierarchy, 

356: like the usual strong hierarchy among the fundamental parameters, namely between the NMSSM 

357: parameters ($\lambda$ or $\mu$, as we will see later) and the \rpv coupling constants 

358: ($\lambda_i$ or $\mu_i$, respectively). The price that

359: one must pay here in order to suppress the neutrino mass is a certain fine tuning on some NMSSM

360: parameters. However, this fine tuning can be greatly softened in specific regions of parameter 

361: space, since several NMSSM parameters enter the effective neutrino mass expression and some of them 

362: through power-law dependence.  

363:   

364: There exist mainly two supersymmetric alternatives to our scenario: two 

365: other kinds of model \cite{AltI,AltII} have been suggested in order to address simultaneously the 

366: hierarchy questions of the $\mu$ naturalness {\it and} the neutrino lightness. We will compare the 

367: characteristics and numerical aspects of these two models with those of our scenario.

368: 

369: In next section, we study the simplest version of our scenario, taking into account 

370: the constraints on NMSSM parameter space issued from collider physics. For that purpose, 

371: we use the NMHDECAY program \cite{nmhdecay}. In Section \ref{versionII},

372: we discuss a version where the bilinear \rpv terms are generated via the spontaneous breaking

373: of a symmetry. Finally, we conclude in Section \ref{conclu}.

374: 

375: 

376: \section{Scenario I}

377: \label{versionI}

378: 

379: \subsection{Neutralino masses}

380: 

381: {\bf $\bullet$ Superpotential:}

382: The superpotential of the NMSSM contains two characteristic terms in addition of the Yukawa 

383: couplings:

384: \begin{equation}

385: W_{NMSSM}= Y^u_{ij} Q_i H_u U_j^c + Y^d_{ij} Q_i H_d D_j^c + Y^\ell_{ij} L_i H_d E_j^c 

386: + \lambda S H_u H_d + \frac{1}{3} \kappa S^3 ,

387: \label{WNMSSM}

388: \end{equation}

389: $Y^{u,d,\ell}_{ij}$ being the Yukawa coupling constants ($i,j,k$ are flavor indexes), 

390: $\lambda$ and $\kappa$ dimensionless coupling constants and $Q_i$, $L_i$, $U^c_i$,

391: $D^c_i$, $E^c_i$, $H_u$, $H_d$, $S$ respectively the superfields for the quark 

392: doublets, lepton doublets, up-type anti-quarks, down-type anti-quarks, anti-leptons, 

393: up Higgs, down Higgs, extra singlet under the SM gauge group $SU(3)_c \times SU(2)_L 

394: \times U(1)_Y$. The $SU(2)_L$ product of the two Higgs doublets $H_d^T=(H_d^0,H_d^-)$ 

395: and $H_u^T=(H_u^+,H_u^0)$ is defined as,

396: \begin{equation}

397: H_u H_d=H_u^+ H_d^- - H_u^0 H_d^0.

398: \label{SU2prod}

399: \end{equation}

400: The absence of terms $H_u H_d$ as well as $S^2$ and tadpoles is insured by a suitable 

401: discrete symmetry. An effective $\mu$ term, $\lambda \langle s \rangle  H_u H_d$, is generated  

402: via a VEV for the scalar component $s$ of the singlet superfield $S$.

403: 

404: In addition to the above NMSSM superpotential, we first consider the bilinear \rpv 

405: interactions:

406: \begin{equation}

407: W_I = W_{NMSSM} + \mu_i L_i H_u ,

408: \label{WmI}

409: \end{equation}

410: where $\mu_i$ are dimension-one \rpv parameters. The presence of the other renormalizable

411: \rpv interactions, namely the trilinear \rpv interactions such as $\lambda_{i,j,k} L_i L_j E_k^c$, 

412: depends on the symmetries of the superpotential that one assumes. It is desirable that the 

413: superpotential symmetries forbid the trilinear \rpv interactions violating either the lepton 

414: or baryon number, or both (as does the R-parity symmetry for example), in order to guarantee 

415: the proton stability \cite{DreinerRoss}. In other words, the whole superpotential symmetry should 

416: be either a Generalized Lepton (GLP), Baryon (GBP) or Matter (GMP) Parity. In case where some 

417: trilinear \rpv interactions are effectively present in the theory with significant coupling constant

418: values, those can possibly induce direct contributions to neutrino masses through one-loop level 

419: diagrams involving squarks or sleptons \cite{AlignI,LoopB}

420: \footnote{The trilinear \rpv couplings can also induce effective masses for neutrinos propagating

421: in matter, via tree level squark or slepton exchanges, but the SNO results forbid these contributions 

422: to be dominant \cite{MoDrei}.}.

423: 

424: 

425: {\bf $\bullet$ Soft terms:}

426: In the soft supersymmetry-breaking part of the Lagrangian, 

427: there exist also \rpv terms \cite{Chemtob}. In the 

428: presence of bilinear \rpv soft terms, the electroweak 

429: symmetry breaking can lead to a non-vanishing VEV for 

430: sneutrinos, denoted $\langle \tilde \nu_i  \rangle $ and 

431: corresponding to a possible spontaneous breaking of R-parity. These 

432: VEV produce new mixings between the neutrinos and neutralinos, 

433: contributing then to Majorana neutrino masses. 

434: \\ However, the $H_d$ and $L_i$ superfields, having identical quantum numbers, can be redefined by an 

435: $SU(4)$ rotation on $(H_d,L_i)^T$. Under this transformation, the \rpv parameters are modified. It is 

436: always possible to find a basis in which either $\langle \tilde \nu_i  \rangle =0$ or $\mu_i=0$. 

437: Nevertheless, 

438: generally $\langle \tilde \nu_i  \rangle $ and $\mu_i$ do not vanish simultaneously 

439: \cite{PhysRep}. In the present

440: framework, we will consider a basis where $\langle \tilde \nu_i  \rangle =0$ and $\mu_i \neq 0$. 

441: 

442: 

443: {\bf $\bullet$ Mass matrix:}

444: Therefore, within our framework, the neutralino mass terms read as,

445: \bea

446: {\cal L}^m_{\tilde \chi^0}=-\frac{1}{2} \Psi^{0^T} {\cal M}_{\tilde \chi^0} \Psi^0 + H.c. 

447: \label{LAGmass}

448: \eea

449: in the basis defined by $\Psi^{0^T}$ $\equiv$ 

450: $(\tilde B^0, \tilde W^0_3, \tilde h^0_d, \tilde h^0_u, \tilde s, \nu_i)^T$,

451: where $\tilde h^0_{u,d}$ ($\tilde s$) is the fermionic component of the superfield $H_{u,d}^0$ ($S$)

452: and the $\nu_i$ denote the neutrinos. 

453: In Eq.(\ref{LAGmass}), the neutralino mass matrix is given by, 

454: \bea

455: {\cal M}_{\tilde \chi^0} = 

456: \left( 

457: \begin{array}{cc}

458: {\cal M}_{NMSSM}  & \xi_{\rpv}^{T} \\

459: \xi_{\rpv} & {\bf 0}_{3 \times 3}

460: \end{array}

461: \right)

462: \label{CHImass}

463: \eea

464: where ${\cal M}_{NMSSM}$ is the neutralino mass matrix which 

465: holds in the NMSSM with conserved R-parity ($s$, $c$ standing for $\sin$, $\cos$): 

466: \bea

467: {\cal M}_{NMSSM}

468: =

469:  \left(

470:  \begin{array}{cccccc}

471: M_{1}&0 & -M_Z \ s\theta_W \ c\beta & M_Z \ s\theta_W \ s\beta & 0\\ 

472: 0 & M_{2} & M_Z \ c\theta_W \ c\beta & -M_Z \ c\theta_W \ s\beta  &0\\

473:  -M_Z \ s\theta_W \ c\beta & M_Z \ c\theta_W \ c\beta & 0 &-\mu& -\lambda v_u\\

474: M_Z \ s\theta_W \ s\beta &-M_Z \ c\theta_W \ s\beta &-\mu&0& -\lambda v_d \\

475: 0&0&-\lambda v_u &-\lambda v_d & 2 \kappa \langle s \rangle  \\

476:   \end{array}

477:  \right)\ ,

478: \label{NMSSMmass}

479: \eea

480: $\xi_{\rpv}$ is the \rpv part of the matrix mixing neutrinos and neutralinos,

481: \bea

482: \xi_{\rpv} = 

483: \left( 

484: \begin{array}{ccccc}

485: 0 & 0 & 0 & \mu_1 & 0  \\

486: 0 & 0 & 0 & \mu_2 & 0  \\

487: 0 & 0 & 0 & \mu_3 & 0  

488: \end{array}

489: \right)

490: \label{RPVmass}

491: \eea

492: $M_1$ ($M_2$) is the soft supersymmetry breaking mass of the bino (wino), $M_Z$ the $Z^0$ boson 

493: mass, $\theta_W$ the electroweak angle, $\tan \beta=v_u/v_d=\langle h^0_u \rangle /\langle h^0_d \rangle $ ($h^0_{u,d}$ being the scalar 

494: component of $H_{u,d}^0$) and 

495: \begin{equation}

496: \mu = \lambda \langle s \rangle .

497: \label{mu}

498: \end{equation}

499: 

500: 

501: {\bf $\bullet$ Parameters:}

502: In this scenario, the independent parameters in the neutralino sector can be chosen

503: as being the following set of variables, 

504: \begin{equation}

505: \lambda, \ \kappa, \ \tan \beta, \ \mu, \ M_1, \ M_2.

506: \label{parameters}

507: \end{equation}

508: We take these variables as free parameters at the electroweak scale. We adopt the convention of signs

509: in which $\lambda > 0$, $\tan \beta > 0$ (without loss of generality) whereas $\kappa$ and $\mu$ can 

510: take positive or negative values. Finally, we assume that $\lambda$, $\kappa$ and the soft supersymmetry 

511: breaking parameters are real.

512: 

513: 

514: \subsection{Effective neutrino mass}

515: \label{effnumass}

516: 

517: 

518: {\bf $\bullet$ Mass expression:}

519: We restrict ourselves to the case $|\mu_i/\mu|<10^{-1}$ and to some parameter values (in particular 

520: sufficiently large $M_{1,2}$) such that the neutrino-neutralino mixing terms remain much smaller 

521: than the neutralino masses. Hence, the effective neutrino mass matrix is given in a good 

522: approximation by the following formula, having a ``see-saw'' type structure,

523: \begin{equation}

524: m_{\nu} = - \xi_{\rpv} \ {\cal M}_{NMSSM}^{-1} \ \xi_{\rpv}^T.

525: \label{seesaw}

526: \end{equation}

527: We have checked, through a comparison with an exact numerical diagonalization, that this block 

528: form expression represents systematically a good approximation for all the points of parameter 

529: space that we consider in this work. From Eq.(\ref{NMSSMmass}), Eq.(\ref{RPVmass}) and 

530: Eq.(\ref{seesaw}), we deduce an analytic expression for the effective Majorana neutrino mass 

531: matrix:

532: \begin{equation}

533: m_{\nu i j} = \mu_i \mu_j \frac{M_1 M_2 (2 \kappa \mu / \lambda)}{Det({\cal M}_{NMSSM})}

534: \bigg (

535: \frac{(\lambda v_u)^2}{2 \kappa \mu / \lambda} + M_Z^2

536: [ \frac{\sin^2\theta_W}{M_1} + \frac{\cos^2\theta_W}{M_2} ] \cos^2 \beta

537: \bigg ),

538: \label{MnuEFF}

539: \end{equation}

540: $Det({\cal M}_{NMSSM})$ being the determinant of the matrix (\ref{NMSSMmass}).

541: 

542: {\bf $\bullet$ Origin of smallness:}

543: We observe on neutrino mass matrix (\ref{MnuEFF}) that the overall factor can be significantly

544: suppressed if the two terms in brackets compensate each other by taking opposite signs and

545: approximately equal absolute values. This means that neutrino mass eigenvalues can be affected 

546: by an important suppression factor. This neutrino mass suppression has a different and new origin 

547: with respect to the other possible suppression coming from the smallness of ratio $|\mu_i/\mu|$ 

548: ({\it c.f.} Eq.(\ref{MnuEFF})). The smallness of ratio $|\mu_i/\mu|$ cannot constitute a physical 

549: interpretation to the smallness of neutrino mass scale compared to electroweak scale, 

550: in the sense that an other (unexplained) mass hierarchy is introduced. 

551: 

552: 

553: \begin{figure}[t]\unitlength1mm

554: \SetScale{2.8}

555: \begin{boldmath}

556: %figure

557: \begin{tabular}{c c c}

558: \begin{picture}(80,80)(0,-10)

559: %\ArrowLine(0,0)(15,0)

560: \Line(0,0)(15,0)\Line(65,0)(15,0)

561: %\ArrowLine(80,0)(65,0)

562: \Line(80,0)(65,0)

563: \DashLine(25,0)(25,30){2}

564: \Text(22,-4.5)[c]{$\widetilde  {h}^0_{u,d}$}

565: \DashLine(55,0)(55,30){2}

566: \Text(60,-4.5)[c]{$\widetilde  {h}^0_{u,d}$}

567: \Text(52.4,29.6)[l]{$\times$}

568: \Text(55,35)[c]{$\langle {h}^0_{u,d} \rangle $}

569: \Text(23,29.6)[l]{$\times$}

570: \Text(25,35)[c]{$\langle {h}^0_{u,d} \rangle $}

571: \Text(73,0)[r]{$<$}

572: \Text(61,0)[r]{$>$}

573: \Text(-2,0)[r]{$\nu_i$}

574: \Text(8,0)[r]{$>$}

575: \Text(22,0)[r]{$<$}

576: \Text(80,0)[l]{$\nu_j$}

577: \Text(40,-6)[c]{$\widetilde{B^0},\widetilde{W}^0_3$}

578: \Text(40,-12)[c]{(a)}

579: \Text(40,5)[c]{$M_{1,2}$}

580: \Text(33,0)[r]{$>$}

581: \Text(50,0)[r]{$<$}

582: \Text(40,0)[c]{$\times$}

583: \Text(15,0)[c]{x}

584: \Text(65,0)[c]{x}

585: \Text(15,5)[c]{$\mu_i$}

586: \Text(65,5)[c]{$\mu_j$}

587: \end{picture}

588: &\strut\hspace*{2mm}&

589: %\ArrowLine(0,0)(15,0)

590: \begin{picture}(80,80)(0,-10)

591: \Line(0,0)(15,0)\Line(65,0)(15,0)

592: %\ArrowLine(80,0)(65,0)

593: \Line(78,0)(65,0)

594: \DashLine(25,0)(25,30){2}

595: \Text(22,-4.5)[c]{$\widetilde  {h}^0_{u,d}$}

596: \DashLine(55,0)(55,30){2}

597: \Text(60,-4.5)[c]{$\widetilde  {h}^0_{u,d}$}

598: \Text(52.6,29.6)[l]{$\times$}

599: \Text(55,35)[c]{$\langle {h}^0_{u,d} \rangle $}

600: \Text(23.2,29.6)[l]{$\times$}

601: \Text(25,35)[c]{$\langle {h}^0_{u,d} \rangle $}

602: \Text(73,0)[r]{$<$}

603: \Text(61,0)[r]{$>$}

604: \Text(-1,0)[r]{$\nu_i$}

605: \Text(8,0)[r]{$>$}

606: \Text(22,0)[r]{$<$}

607: \Text(80,0)[l]{$\nu_j$}

608: \Text(40,-6)[c]{$\widetilde{s}$}

609: \Text(33,0)[r]{$>$}

610: \Text(50,0)[r]{$<$}

611: \Text(40,5)[c]{$2\kappa \mu/\lambda$}

612: \Text(40,0)[c]{$\times$}

613: \Text(15,0)[c]{x}

614: \Text(65,0)[c]{x}

615: \Text(15,5)[c]{$\mu_i$}

616: \Text(62,5)[c]{$\mu_j$}

617: \Text(40,-12)[c]{(b)}

618: \end{picture}

619: \end{tabular}

620: \end{boldmath}

621: \caption{(a) Feynman diagram for the typical contribution to the Majorana neutrino masses arising 

622: in the MSSM from mixing with neutralinos (see text for notations of fields and parameters).

623: The effective mass affecting the two vertex is of type $m=\pm M_Z t(\theta_W) t(\beta)$, 

624: where $t(x)$ is equal to either $\sin x$ or $\cos x$. 

625: A cross indicates either a mass insertion or 

626: a VEV. The arrows show the flow of momentum for associated propagators.

627: (b) Feynman diagram for the additional type of contribution to the Majorana neutrino masses 

628: arising in the NMSSM from mixing with neutralinos. The mass parameter at the two vertex 

629: is there $m= - \lambda v_{u,d}$.}

630: \protect\label{fig:diagramMSSM}

631: \end{figure}

632: 

633: 

634: Let us understand this possible cancellation in Eq.(\ref{MnuEFF}) 

635: from a diagramatic point of view. In Fig.(\ref{fig:diagramMSSM}), 

636: we present the two characteristic diagrams of neutralino mass 

637: contributions to Majorana neutrino mass terms. We see that the first term in brackets of neutrino 

638: mass expression (\ref{MnuEFF}) corresponds to the exchange of a gaugino shown in Fig.(\ref{fig:diagramMSSM})(a). 

639: Indeed, this first term is of the type $\mu_i \mu_j (m^2 / M)$, where $m$ is the typical mass entering 

640: at the two vertex linked to a Higgs VEV and $M=M_{1,2}$ is the gaugino   

641: mass. The second term in brackets of formula (\ref{MnuEFF}) is associated to the exchange of the singlino 

642: $\tilde s$ shown in Fig.(\ref{fig:diagramMSSM})(b): this term is also of the type $\mu_i \mu_j (m^2 / M)$, 

643: where $m$ is the typical mass at vertex with a Higgs VEV and now 

644: $M= 2 \kappa \langle s \rangle  = 2 \kappa \mu / \lambda$ is the singlino mass 

645: (see Eq.(\ref{NMSSMmass}) and Eq.(\ref{mu})).

646: In conclusion, an approximate cancellation between the two terms in the brackets entering neutrino mass 

647: expression (\ref{MnuEFF}) would represent a compensation between the exchanges of a gaugino and a singlino.

648: 

649: 

650: The price of this neutrino mass suppression 

651: is a certain amount of fine tuning on some NMSSM parameter values. The $\lambda$ parameter faces the most 

652: important fine tuning. Nevertheless, this fine tuning is significantly reduced when $| \mu |$, 

653: $M_{1,2}$ and $\tan \beta$ increase. In next section, we discuss this aspect more precisely 

654: and quantitatively.

655: 

656: 

657: \subsection{Numerical results}

658: \label{NumResI}

659: 

660: {\bf $\bullet$ Flavors:}

661: In the discussion of the main features of our scenario, we will concentrate on the 

662: case of one neutrino flavor, for simplification reasons. The treatment of the realistic 

663: three-flavor case requires the calculation of loop contributions to the neutrino mass

664: matrix, via the Grossman-Haber diagrams \cite{GroHab} which kill the 

665: degeneracy in neutrino mass spectrum. 

666: Indeed, at the tree level, only one of the three neutrino eigenstates 

667: obtains a non-vanishing mass eigenvalue (from the mixing with neutralinos), a scheme which 

668: conflicts with present data as we know that solar and atmospheric neutrino data

669: require at least two non-zero eigenvalues \cite{Valle}. The possibility, that the combined tree and 

670: one-loop contributions could account for the observed data on three neutrino masses and three

671: leptonic mixing angles, has been investigated extensively in the context of the MSSM 

672: \cite{MSSMrpvA}-\cite{MSSMrpvB}.

673: Such a three-flavor global fit of all neutrino data at the loop level within the NMSSM

674: is beyond the scope of our study. Nevertheless, we comment that, as in the MSSM, the soft supersymmetry

675: breaking interactions (like $B_i h^0_u \tilde \nu_i$) \cite{GroHab}, the cancellations between 

676: contributions involving the Higgs sector modes \cite{cancel} as well as the sneutrino mass splittings 

677: should play a crucial r\^ole in the computations for loop amplitudes of neutrino masses.

678: 

679: At one flavor, we see from Eq.(\ref{MnuEFF}) that the neutrino mass can be written as, 

680: \begin{equation}

681: m_{\nu} = \frac{\mu_1^2}{M_{SUSY}},

682: \label{MnuOne}

683: \end{equation}

684: where $M_{SUSY}$ is an effective mass depending on the supersymmetry (breaking) parameters.

685: In the three-flavor case, the only non-vanishing neutrino mass eigenvalue at tree level reads as,

686: \begin{equation}

687: m^{high}_{\nu} = \frac{\mu_1^2+\mu_2^2+\mu_3^2}{M_{SUSY}}.

688: \label{MnuThree}

689: \end{equation}

690: At loop level, the two other neutrino mass eigenvalues receive loop contributions. 

691: We consider that the largest neutrino mass eigenvalue remains given by $m^{high}_{\nu}$ 

692: in Eq.(\ref{MnuThree}) to a good approximation. By consequence, the neutrino mass $m_{\nu}$ in 

693: Eq.(\ref{MnuOne}) that we will consider at one flavor, is approximately equal to the largest 

694: neutrino mass eigenvalue at three flavors, namely $m^{high}_{\nu}$ in Eq.(\ref{MnuThree}), 

695: for $\mu_{2,3}^2 \lesssim \mu_1^2$.

696: 

697: 

698: {\bf $\bullet$ Neutrino mass constraints:}

699: Let us summarize the existing experimental constraints on the largest neutrino mass eigenvalue

700: $m^{high}_{\nu}$.

701: First, a three-flavor global fit analysis, including the results from solar, atmospheric, reactor (KamLAND 

702: and CHOOZ) and accelerator (K2K) experiments, leads to the following intervals at the $4 \sigma$ 

703: level \cite{Valle}:

704: $6.8 \leq \Delta m_{21}^2 \leq 9.3 \ \ \ [10^{-5} \mbox{eV}^2]$ and

705: $1.1 \leq \Delta m_{31}^2 \leq 3.7 \ \ \ [10^{-3} \mbox{eV}^2]$,  

706: $\Delta m_{21}^2 \equiv m_{\nu_2}^2-m_{\nu_1}^2$ and $\Delta m_{31}^2 \equiv m_{\nu_3}^2-m_{\nu_1}^2$ 

707: being the differences of squared neutrino mass eigenvalues. Hence, the largest neutrino mass eigenvalue

708: is larger than about $\sqrt{3 \ 10^{-3} \mbox{eV}^2}$, which can be written as,

709: \begin{equation}

710: 0.05 \mbox{eV} \lesssim m^{high}_{\nu}.

711: \label{bdINF}

712: \end{equation} 

713: We now turn to the current upper experimental limits on absolute neutrino mass scales. We first

714: consider the limits extracted from the tritium beta decay experiments \cite{107,108,109} which

715: are independent of the nature of neutrino mass (Majorana or Dirac). The data provided by the 

716: Mainz \cite{108} and Troitsk \cite{109} experiments give rise to the bounds (at $95 \% \ C.L.$):

717: $m_\beta \leq 2.2 \ \mbox{eV} \ \ \ \mbox{[Mainz]}$ and

718: $m_\beta \leq 2.5 \ \mbox{eV} \ \ \ \mbox{[Troitsk]}$,

719: where the effective mass $m_\beta$ is defined by $m^2_\beta = \sum_{i=1}^3 |U_{ei}|^2 m_{\nu_i}^2$,

720: $U_{ei}$ being the leptonic mixing matrix. This matrix is parameterized 

721: by the three mixing angles $\theta_{12}$, $\theta_{23}$ and $\theta_{13}$ 

722: which are constrained to lie in the ranges \cite{Valle}:

723: $0.21 \leq \sin^2 \theta_{12} \leq 0.41$, 

724: $0.30 \leq \sin^2 \theta_{23} \leq 0.72$ and

725: $\sin^2 \theta_{13} \leq 0.073$. 

726: From the above constraints, we deduce that the largest neutrino 

727: mass eigenvalue is bounded from above typically by

728: \begin{equation}

729: m^{high}_{\nu} \lesssim 1 \mbox{eV}.

730: \label{bdSUP}

731: \end{equation} 

732: Secondly, the cosmological data from WMAP and 2dFGRS galaxy survey \cite{cosmobound} place the

733: following bound (depending on cosmological priors):

734: $\sum_{i=1}^3 m_{\nu_i} \lesssim 0.7 \mbox{eV}$.

735: This bound gives rise to an upper limit on the largest 

736: neutrino mass eigenvalue which is of the same order of magnitude as in Eq.(\ref{bdSUP}).

737: 

738: 

739: As we have discussed above, the largest neutrino mass eigenvalue $m^{high}_{\nu}$, 

740: at three flavors, is approximately equal to the neutrino mass $m_{\nu}$, at one flavor.

741: One thus concludes from the typical bounds (\ref{bdINF}) and (\ref{bdSUP}) that

742: \begin{equation}

743: 0.1 \mbox{eV} \lesssim m_{\nu} \lesssim 1 \mbox{eV}. 

744: \label{bdTOT}

745: \end{equation} 

746: 

747: \begin{table}[t]

748: \begin{center}

749: \begin{tabular}{c|c|c|c|c|c|c|c}

750:  & $\kappa$ & $\mu$ & $\tan \beta$ & $M_1$ & $M_2$ & $\mu_1$ & $\lambda$  

751: \\

752: &  & $\mbox{[GeV]}$ &  & $\mbox{[TeV]}$ & $\mbox{[TeV]}$ & $\mbox{[GeV]}$ &

753: \\

754: \hline

755: A & 0.05 &  -400 &  54   & 5 & 5 & $10$ & $(9.0969 - 9.105) \ 10^{-3}$ 

756: \\

757: &  &   &    &  &  & $10^{-1}$ & $(1.43 - 2.8) \ 10^{-2}$ 

758: \\

759: \hline

760: B & 0.05 &  -300 &  50   & 1 & 1 & $10$ & $(1.48759 - 1.48771) \ 10^{-2}$ 

761: \\

762: &  &   &   & & & $10^{-1}$ & $(1.613 - 2.31) \ 10^{-2}$ 

763: \\

764: \hline

765: C & 0.15 &  -300 &  30   & 5 & 5 & $10$ & $(1.76374 - 1.764) \ 10^{-2}$ 

766: \\

767: & &  &   &  &  & $10^{-1}$ & $(2.014 - 3.2) \ 10^{-2}$ 

768: \\

769: \hline

770: D & 0.2 &  -200 &  50   & 3 & 4 & $10$ & $(1.3321 - 1.3323) \ 10^{-2}$ 

771: \\

772: & &  &   &  &  & $10^{-1}$ & $(1.51 - 2.35) \ 10^{-2}$ 

773: \\

774: \hline

775: E & -0.1 &  300 &  50   & 3 & 3 & $10$ & $(1.29955 - 1.2999) \ 10^{-2}$ 

776: \\

777: &  &  &   &  &  & $10^{-1}$ & $(1.585 - 2.72) \ 10^{-2}$ 

778: \end{tabular}

779: \caption{Sets (A,\dots,E) of values, for the parameters entering the whole neutralino mass matrix 

780: (\ref{CHImass}), which reproduce the correct neutrino mass. The two values of parameter $\lambda$ 

781: correspond to the neutrino masses $m_{\nu}= 0.1 \mbox{eV} - 1 \mbox{eV}$ ($m_{\nu}$ being defined 

782: via Eq.(\ref{MnuEFF})), respectively. As the one-flavor case is considered here, the flavor index

783: $i$ of \rpv parameter $\mu_i$ takes only the value $i=1$ (as in Eq.(\ref{MnuOne})).}

784: \label{tab:parI}

785: \end{center}

786: \end{table}

787: 

788: 

789: {\bf $\bullet$ Neutrino mass suppression:}

790: In Table \ref{tab:parI}, we present characteristic points of parameter space for which

791: the neutrino mass (\ref{MnuEFF}) at one flavor, namely $m_{\nu}$ ({\it c.f.} Eq.(\ref{MnuOne})), 

792: is equal to $0.1 \mbox{eV}$ and $1 \mbox{eV}$, in order to cover the typical range of values 

793: allowed by experimental results (see Eq.(\ref{bdTOT})).

794: 

795: In fact, for each of the sets of parameters shown in Table \ref{tab:parI}, the $\lambda$ value

796: is determined as a function of the other parameters through the formula (\ref{MnuEFF}) for neutrino 

797: mass. In other terms, the relation (\ref{MnuEFF}) fixes one of the parameters (as the $m_{\nu}$ 

798: value is given) that we choose to be $\lambda$. The $\lambda$ values are written with the accuracy

799: necessary to obtain the wanted neutrino mass. This accuracy reflects two aspects: the fact that it is 

800: $\lambda$ that we determine as a function of the other parameters, and, the fine tuning needed on 

801: $\lambda$ (which will be discussed in more details in next table). As already said, $\lambda$

802: is the quantity that suffers from the most important fine tuning. 

803: 

804: 

805: Let us discuss the physical meaning of results presented in Table \ref{tab:parI}. 

806: We remark that for the signs of parameters systematically chosen in this table

807: (note the different sign configuration for last point E), the approximate cancellation 

808: between the two terms in brackets entering neutrino mass expression (\ref{MnuEFF}) 

809: is effective as these two terms possess opposite signs. The first possibility is that this 

810: cancellation is only partially responsible for the neutrino mass suppression relatively to 

811: the electroweak scale: this is the case for all the points in this table with 

812: $| \mu_1/\mu | \simeq 10^{-3}$ ($\mu_1 = 10^{-1} \mbox{GeV}$). In that case, the 

813: suppression of neutrino mass is also due to the hierarchy introduced between the \rpv 

814: parameter $\mu_1$ and the effective $\mu$ quantity. The other possibility is that the above

815: cancellation constitutes the main mechanism suppressing the neutrino mass: this is the case 

816: for the points with $| \mu_1/\mu | \simeq 10^{-1}$ ($\mu_1 = 10 \mbox{GeV}$). In that case, 

817: the necessary neutrino mass suppression is achieved without introducing any new strong 

818: hierarchy among the parameters of the theory. 

819: 

820: This result, that the smallness of neutrino mass can be mainly

821: due to a compensation between two contributions exchanging a gaugino and a singlino, is 

822: one of the major and new results of our paper.

823: 

824: 

825: {\bf $\bullet$ $\mu$ naturalness:}

826: Let us comment about the parameter values taken in Table \ref{tab:parI}. Motivated by arguments of 

827: naturalness, one may wish to restrict to $\langle s \rangle  \lesssim 10 \mbox{TeV}$, which translates ({\it c.f.} 

828: Eq.(\ref{mu})) into the condition $|\mu| [\mbox{GeV}] \times 10^{-4} \lesssim \lambda$. This condition 

829: is satisfied by the values obtained in Table \ref{tab:parI}. Besides, the absence of Landau singularities,

830: for $\lambda$, $\kappa$ and the Yukawa coupling constants $Y^b,Y^t$ below the GUT energy scale, imposes 

831: \cite{darkNMSSM} the typical bounds on NMSSM parameters: $\lambda \lesssim 0.75$, $|\kappa| \lesssim 0.65$ 

832: and $1.7 \lesssim \tan \beta \lesssim 54$. All the parameter values in Table \ref{tab:parI} satisfy 

833: these bounds. Finally, the various values of $\mu$ in this table have been chosen such that 

834: $|\mu| \gtrsim 100 \mbox{GeV}$, in order to safely respect the LEP bound on the lightest chargino 

835: mass: $m_{\tilde \chi_1^+}>103.5 \mbox{GeV}$ \cite{LEPchargino}.

836: 

837: 

838: In addition, we have checked that the parameter sets presented in Table \ref{tab:parI} belong well

839: to some regions of the NMSSM parameter space which are compatible with the various theoretical consistencies 

840: and experimental constraints. For that purpose, we have performed a scan, by using the Fortran code NMHDECAY 

841: \cite{nmhdecay}, in order to test the following parameter ranges: $0.009< \lambda <0.02$, $0.05< | \kappa | <0.2$,

842: $30< \tan \beta <54$ and $100\mbox{GeV}< | \mu | <400\mbox{GeV}$. This scan was done simultaneously with a scan 

843: over $-1\mbox{TeV}< A_\lambda <1\mbox{TeV}$ and $-1\mbox{TeV}< A_\kappa <1\mbox{TeV}$, where $A_\lambda$ and 

844: $A_\kappa$ are the trilinear soft supersymmetry breaking parameters (entering the NMSSM Lagrangian via the terms 

845: $\lambda A_\lambda s h_u h_d$ and $(1/3) \kappa A_\kappa s^3$) which do not affect the neutralino mass matrix 

846: (\ref{CHImass}). Precisely, 

847: the NMHDECAY program has allowed us to check that \cite{nmhdecay} {\it (i)} the physical 

848: minimum of the scalar potential is deeper than the local unphysical minima with $\langle h^0_{u,d} \rangle =0$ and/or $\langle s \rangle =0$

849: {\it (ii)} the running couplings $\lambda$, $\kappa$, $Y^b$ and $Y^t$ do not encounter a Landau singularity 

850: {\it (iii)} the experimental constraints from LEP in the neutralino, chargino and Higgs sectors are 

851: effectively satisfied. 

852: 

853: 

854: The consistency of using the code NMHDECAY (which strictly speaking deals with the pure NMSSM) in our present 

855: scenario is justified by the following argument.

856: The presence of the additional bilinear \rpv term $\mu_i L_i H_u$ in the superpotential (see Eq.(\ref{WmI})), 

857: that we have supposed, does not automatically modifies the Higgs potential of the NMSSM at tree level. 

858: Indeed, the term $\mu_i^2 |h_u|^2$ in the Higgs potential, coming from the bilinear \rpv term, can be 

859: reabsorbed in a redefinition of the soft Higgs mass term $m_{h_u}^2 |h_u|^2$.

860: 

861: 

862: \begin{table}[t]

863: \begin{center}

864: \begin{tabular}{c|c|c|c}

865: & $m_{\tilde \chi^0_1}$  & $m_{\tilde \chi^0_5}$  &  ${\cal F}_\lambda$  

866: \\

867: & $\mbox{[GeV]}$ & $\mbox{[GeV]}$ &

868: \\

869: \hline

870: A & 399 & 5002 &   $(0.9 - 9.3) \ 10^{-4}$ 

871: \\

872: & 399 & 5002 & $(2.5 - 3.2) \ 10^{-1}$   

873: \\

874: \hline

875: B & 295 & 2017 &  $(0.9 - 9.0) \ 10^{-5}$ 

876: \\

877: & 294 & 1862 - 1303  & $(0.7 - 2.4) \ 10^{-1}$   

878: \\

879: \hline

880: C & 299 & 5103 - 5102  &  $(0.2 - 1.6) \ 10^{-4}$ 

881: \\

882: & 299 & 5002 & $(1.1 - 2.8) \ 10^{-1}$ 

883: \\

884: \hline

885: D & 199 & 6006 - 6005 &  $(0.1 - 1.5) \ 10^{-4}$ 

886: \\

887: & 199 & 5310 - 4002 &  $(1.0 - 2.7) \ 10^{-1}$ 

888: \\

889: \hline

890: E & 299 & 4617 - 4616  & $(0.3 - 2.7) \ 10^{-4}$ 

891: \\

892: & 298 & 3787 - 3003 &   $(1.5 - 3.0) \ 10^{-1}$ 

893: \end{tabular}

894: \caption{Lowest [$m_{\tilde \chi_1^0}$] and highest [$m_{\tilde \chi_5^0}$] neutralino masses 

895: (among the six mass eigenvalues of matrix (\ref{CHImass}), 

896: except the neutrino mass eigenvalue $m_{\nu}$) for the points A,\dots,E of parameter space presented in 

897: Table \ref{tab:parI}. Together with these masses, we also show the value of fine tuning function 

898: ${\cal F}_\lambda$ defined in the text for the $\lambda$ parameter. The two values of ($m_{\tilde \chi_5^0}$ 

899: and) ${\cal F}_\lambda$ correspond respectively to the two $\lambda$ values in Table \ref{tab:parI}

900: (leading to $m_{\nu}= 0.1 \mbox{eV} - 1 \mbox{eV}$). For each point, the first and second lines are 

901: respectively associated to $\mu_1=10 \mbox{GeV}$ and $\mu_1=10^{-1} \mbox{GeV}$, as in Table 

902: \ref{tab:parI}.}

903: \label{tab:FTI}

904: \end{center}

905: \end{table}

906: 

907: 

908: 

909: {\bf $\bullet$ Fine tuning:}

910: The mechanism of neutrino mass suppression presented in Section \ref{effnumass} requires a certain

911: amount of fine tuning. In order to discuss quantitatively this fine tuning on the $\lambda$ parameter 

912: (the most important fine tuning), we introduce the following ratio,

913: \begin{equation}

914: {\cal F}_\lambda 

915: = \bigg | \frac{\delta ln \lambda}{\delta ln m_{\nu}} \bigg | 

916: = \bigg | \frac{\delta \lambda / \lambda}{\delta m_{\nu} / m_{\nu}} \bigg | ,

917: \label{FTratio}

918: \end{equation} 

919: where $\delta m_{\nu}$ is the variation of neutrino mass associated to the variation $\delta \lambda$ 

920: of fundamental parameter $\lambda$, for any other parameter fixed to a certain value. The largest values 

921: of this quantity ${\cal F}_\lambda$ correspond to the most soft fine tuning. By using the neutrino mass 

922: expression (\ref{MnuEFF}), we have calculated analytically the quantity ${\cal F}_\lambda$ as a function 

923: of the fundamental parameters of the neutralino mass matrix.

924: 

925: In Table \ref{tab:FTI}, we give the values of this function ${\cal F}_\lambda$ for the points of 

926: parameter space presented in Table \ref{tab:parI} which generate acceptable neutrino masses through 

927: our cancellation mechanism. By comparing the points A and B of Table \ref{tab:FTI}, we observe, through

928: the values of function ${\cal F}_\lambda$, that the fine tuning get softer as $M_{1,2}$ increases. 

929: Similarly, the comparison of parameter sets A and C (A and D) shows that the fine tuning is significantly 

930: improved for larger values of $\tan \beta$ ($| \mu |$). The point E, corresponding to different signs 

931: of $\kappa$ and $\mu$ than for the other points, exhibits the weak dependence of fine tuning on the 

932: sign configurations. Finally, we remark that the fine tuning is softer for an higher neutrino mass

933: (second ${\cal F}_\lambda$ values in Table \ref{tab:FTI}) as well as for a smaller $|\mu_i/\mu|$ ratio

934: (second line for each point). The reason is that, in these two cases, the neutrino mass suppression 

935: mechanism, which is based on the compensation of two mass contributions, has to be less effective

936: (i.e. it must suppresses less the absolute neutrino mass scale).

937: 

938: 

939: To finish the comments about Table \ref{tab:FTI}, we mention that, for each parameter set considered, 

940: the largest neutralino mass eigenvalue $m_{\tilde \chi^0_5}$ is of order of the $\mbox{TeV}$ scale 

941: so that the gauge hierarchy problem remains addressed through the supersymmetry.

942: 

943: \begin{figure}[t]

944: \begin{center} 

945: \psfrag{mu}[c][c][1]{{\large $\mu \ \mbox{[GeV]}$}} 

946: \psfrag{tanbeta}[c][r][1]{{\large $\tan \beta$}}

947: \includegraphics[width=0.5\textwidth,height=6cm]{mutb.eps}

948: \caption{Points in the plan $\mu$ (in $\mbox{GeV}$) versus $\tan \beta$ producing a neutrino mass

949: $m_{\nu}=1 \mbox{eV}$, for $\mu_1 = 1 \mbox{GeV}$, $M_1=3 \mbox{TeV}$, $M_2=4 \mbox{TeV}$ 

950: and values of $\lambda$ and $\kappa$ given by Fig.(\ref{fig:lakp}).} 

951: \protect\label{fig:mutb}

952: \end{center}

953: \end{figure}

954: 

955: \begin{figure}[t]

956: \begin{center} 

957: \psfrag{lambda}[c][c][1]{{\large $\lambda$}} 

958: \psfrag{kappa}[c][r][1]{{\large $\kappa$}}

959: \psfrag{pta}[c][c][1]{{\footnotesize 0.0025}}

960: \psfrag{ptb}[c][c][1]{{\footnotesize 0.005}}

961: \psfrag{ptc}[c][c][1]{{\footnotesize 0.0075}}

962: \psfrag{ptd}[c][c][1]{{\footnotesize 0.01}}

963: \psfrag{pte}[c][c][1]{{\footnotesize 0.0125}}

964: \psfrag{ptf}[c][c][1]{{\footnotesize 0.015}}

965: \psfrag{ptg}[c][c][1]{{\footnotesize 0.0175}}

966: \includegraphics[width=0.6\textwidth,height=6.5cm]{lakp.eps}

967: \caption{Points in the plan $\lambda$ versus $\kappa$ producing a neutrino mass

968: $m_{\nu}=1 \mbox{eV}$, for $\mu_1 = 1 \mbox{GeV}$, $M_1=3 \mbox{TeV}$, $M_2=4 \mbox{TeV}$ 

969: and values of $\mu$ and $\tan \beta$ given by Fig.(\ref{fig:mutb}).}

970: \protect\label{fig:lakp}

971: \end{center}

972: \end{figure}

973: 

974: {\bf $\bullet$ Scans:}

975: In Fig.(\ref{fig:mutb}) and Fig.(\ref{fig:lakp}), we show points of the NMSSM parameter space

976: generating a neutrino mass at one flavor ({\it c.f.} Eq.(\ref{MnuOne})) equal to $1 \mbox{eV}$,

977: for $\mu_1=1\mbox{GeV}$. This $\mu_1$ value corresponds to $| \mu_1/\mu | \simeq 10^{-2}$, 

978: which means that, for the points presented in the two figures, the dominant effective suppression 

979: mechanism of neutrino mass is our cancellation mechanism (a ratio of $|\mu_i/\mu| \sim 10^{-6}$ 

980: is needed to obtain the entire neutrino mass reduction from the hierarchy between $\mu_i$ and $\mu$,

981: as we will discuss later). 

982: 

983: The points in Fig.(\ref{fig:mutb})-(\ref{fig:lakp}) have been obtained through a scan performed 

984: with the NMHDECAY program, so that they respect the experimental and theoretical constraints mentioned 

985: above.

986: 

987: These two figures show that an acceptable neutrino mass can be generated, via the considered

988: cancellation model, in large regions of the NMSSM parameter space. Besides, Fig.(\ref{fig:lakp}) 

989: exhibits a correlation between the coupling constants $\lambda$ and $\kappa$ which is characteristic

990: of the cancellation mechanism.  

991: 

992: 

993: {\bf $\bullet$ Lepton flavor violation:}

994: In the present framework, we have shown that the experimental values of neutrino masses allow the

995: ratio $| \mu_i/\mu |$ to be as large as $\sim 10^{-1}$. Such a possible enhancement of $| \mu_i/\mu |$

996: tends to increase the amplitudes of low energy lepton flavor violating processes like $\mu \to eee$

997: or $\mu \to e \gamma$. Indeed, these decay processes receive tree level contributions through the mixings

998: of type $\mu_i \tilde h^+_u \ell_i$. We obtain, via simple estimations (as done for instance in 

999: \cite{adl}), 

1000: that the experimental upper limits \cite{PDG} on the branching ratios of these decay processes are 

1001: respected for $| \mu_i/\mu |$ values up to $\sim 10^{-2}$.

1002:  

1003: 

1004: \subsection{Comparison with the other models}

1005: 

1006: 

1007: Finally, we compare our scenario with existing alternative supersymmetric models. First, it has been suggested

1008: recently \cite{AltII} that a gauge-singlet right handed neutrino $N_i^c$, added to the MSSM superfield content in 

1009: order to generate Dirac neutrino masses (via $Y^\nu_{ij} L_i H_u N_j^c$), can also play the r\^ole of the NMSSM 

1010: singlet S. Indeed, the scalar components of $N_i^c$ (sneutrinos) can produce an effective $\mu$ term (via 

1011: $\lambda^i N_i^c H_u H_d$) by acquiring a VEV. In this so-called new MSSM, 

1012: the R-parity is broken explicitly via the cubic term  

1013: for $N_i^c$ ($(1/3)\kappa^{ijk}N_i^cN_j^cN_k^c$). The two other model-building differences of this new MSSM 

1014: with our scenario are that, here, the added gauge-singlet $N_i^c$ comes with a flavor index and has a right 

1015: handed chirality (in contrast with $S$). Hence, in the new MSSM, there are three distinct origins to the 

1016: neutrino mass: the mixing with gauginos/higgsinos, the Majorana neutrino mass proportional to $\kappa^{ijk}$ and 

1017: the Dirac neutrino mass involving the Yukawa coupling constants $Y^\nu_{ij}$. These Yukawa coupling constants 

1018: must be of order $10^{-6}$ in order to obtain reasonable neutrino mass eigenvalues around $10^{-2}\mbox{eV}$ 

1019: \cite{AltII}. This means that a hierarchy of $\sim 10^{-6}$ has to be introduced between the Yukawa couplings 

1020: of the neutrinos and the top quark ($Y^t \sim 1$). This has to be contrasted with our mechanism which can produce 

1021: acceptable neutrino masses with only a little hierarchy of $|\mu_i/\mu| \sim 10^{-1}$. Concerning the $\mu$ 

1022: problem, it was shown in \cite{AltII} that the potential minimization conditions are similar to the ones in 

1023: the NMSSM, with the substitution $N^c \leftrightarrow S$. 

1024: 

1025: 

1026: There exist another model \cite{AltI} aimed at solving both the neutrino mass and $\mu$ term problems. 

1027: Within this model, three gauge-singlets are added to the MSSM superfield content: a right handed neutrino

1028: $N_i^c$ giving rise to Dirac neutrino masses, a singlet $S$ addressing the $\mu$ naturalness ``\`a la NMSSM'',

1029: and, a singlet $\Phi_i$ which is essential in order to drive simultaneously a spontaneous breaking of the 

1030: R-parity and electroweak symmetries in a phenomenologically consistent way. In this framework, it is not clear

1031: from the related literature \cite{AltI} what must be the typical neutrino Yukawa coupling values in order to

1032: generate a physical neutrino mass scale around the $\mbox{eV}$.  

1033: 

1034: 

1035: We now compare our scenario, namely the NMSSM in the presence of the \rpv bilinear term $\mu_i L_i H_u$ ({\it c.f.} 

1036: superpotential (\ref{WmI})), with the MSSM in the presence of this same bilinear term. The latter scenario, which 

1037: suffers from the $\mu$ problem, was extensively studied in regard of the neutrino mass aspect \cite{PhysRep}.

1038: \\ Let us consider a generic basis in which $v_i=\langle  \tilde \nu_i  \rangle  \neq  0$ and $\mu_i \neq 0$ simultaneously. 

1039: Then, requiring a neutrino mass scale typically smaller than $1\mbox{eV}$ imposes the alignment \cite{AlignI,AlignII} 

1040: of vectors $v_\alpha \equiv (v_d,v_i)$ and $\mu_\alpha \equiv (\mu,\mu_i)$ up to 

1041: $$

1042: \sin \zeta \lesssim 3 \ 10^{-6} \ \sqrt{1+\tan^2 \beta},

1043: $$ 

1044: where the basis-independent angle $\zeta$ is defined by 

1045: \footnote{For a general discussion on basis-independent parametrization, see \cite{davidson}.},

1046: $$

1047: \cos \zeta = \frac{\sum_\alpha v_\alpha \mu_\alpha}{\sqrt{(\sum_\alpha v_\alpha^2)(\sum_\alpha \mu_\alpha^2)}}.

1048: $$

1049: Such an alignment arises naturally in the framework of horizontal symmetries, but it would then rely on the condition

1050: $|\mu_i| \ll |\mu|$, or more precisely $|\mu_i / \mu| 

1051: < {\cal O}(10^{-5})$ in the first explicit realization proposed in 

1052: \cite{AlignII}. Once more, this hierarchy is more dramatic than in our scenario, where a ratio 

1053: $|\mu_i/\mu| \simeq 10^{-1}$ allows a sufficient neutrino mass suppression relatively to the electroweak energy scale.

1054: \\ Besides, in various accurate three-flavor analyzes \cite{MSSMrpvA}-\cite{MSSMrpvB}, 

1055: it was shown that the combined tree and loop 

1056: MSSM contributions can accommodate the experimental measurements on neutrino masses and leptonic mixing angles. 

1057: In particular, complete scans of the parameter space \cite{AM1,AM2,abl} have shown that the basis-independent

1058: quantities $\delta_\mu^i$ and $\delta_B^i$ (see \cite{DL1}) must be of order 

1059: $|\delta_\mu^i| \sim 10^{-7}$ and $|\delta_B^i| \sim 10^{-5}$, 

1060: assuming sparticle masses fixed at a common effective supersymmetry scale equal to $100\mbox{GeV}$. In the basis 

1061: where $v_i=0$, these two quantities correspond respectively to the ratios $|\mu_i/\mu|$ and $|B_i/B|$

1062: ($\mu_i$ and $B_i$ can be negative), 

1063: $B$ being the soft supersymmetry breaking parameter entering the scalar potential via the interaction 

1064: $B h_u h_d$. So in this basis (that we have considered throughout the study of our scenario), the required ratio 

1065: $|\mu_i/\mu| \sim 10^{-7}$ is much smaller than in our scenario where $|\mu_i/\mu|$ can reach $\sim 10^{-1}$,

1066: with respect to the correct order of magnitude for the neutrino mass scale.

1067: The trilinear \rpv terms, if included, do not change the order of magnitude of the ranges for $\delta_\mu^i$ 

1068: and $\delta_B^i$, and, the \rpv trilinear coupling constants were found to be 

1069: $\sim 10^{-4}$ to satisfy all constraints from neutrino data. 

1070: 

1071: 

1072: 

1073: \section{Scenario II}

1074: \label{versionII}

1075: 

1076: \subsection{Neutralino masses}

1077: 

1078: {\bf $\bullet$ Superpotential:}

1079: We turn to a version of our scenario, proposed in Section \ref{versionI}, where the bilinear 

1080: \rpv interactions have the same origin as the $\mu$ term: those are now generated through the

1081: VEV of the scalar component of the $S$ singlet superfield. Indeed, let us assume that the bilinear 

1082: \rpv interactions of Eq.(\ref{WmI}) are forbidden by a symmetry (exactly like a symmetry is 

1083: imposed within the NMSSM in order to kill the term $\mu H_u H_d$). Then the following 

1084: supersymmetric and gauge invariant term, which is renormalizable, generates $\mu_i$-like terms: 

1085: \begin{equation}

1086: W_{II} = W_{NMSSM} + \lambda_i S L_i H_u,

1087: \label{WmII}

1088: \end{equation}

1089: where $\lambda_i$ are new dimensionless coupling constants. This trilinear 

1090: term, which has no analog in the MSSM, could be rotated away, by an $SU(4)$ rotation on 

1091: $(H_d,L_i)^T$, into the pure NMSSM term $S H_u H_d$. However, the trilinear term $S L_i H_u$

1092: would be regenerated via the renormalization group equations (in the presence of L violating 

1093: couplings) \cite{NMSSM,NMpheno,littleFT}. We also

1094: note that no massless Goldstone boson (the problematic Majoron) appears when $s$ acquires a 

1095: VEV, since the lepton number is already explicitly broken by the trilinear term of Eq.(\ref{WmII}).

1096: This trilinear term also violates explicitly the R-parity symmetry, as the bilinear $\mu_i$ terms 

1097: of superpotential (\ref{WmI}).

1098: The existence of the other trilinear \rpv interactions depends on the superpotential symmetry.

1099: In order to protect the proton against its possible decay channels, this symmetry could be a GLP 

1100: (killing $\lambda_{i,j,k} L_i L_j E_k^c$, $\lambda'_{i,j,k} L_i Q_j D_k^c$ and $\lambda_i S L_i H_u$), 

1101: a GBP (killing $\lambda''_{i,j,k} U_i^c D_j^c D_k^c$) or a GMP (forbidding both L and B violating 

1102: trilinear terms). It is desirable that all the global symmetries of the superpotential are discrete 

1103: gauge symmetries \cite{Ibanez}. Under this hypothesis, by imposing the non-trivial conditions of linear 

1104: anomaly (except the gravitational one) cancellation on the original $Z_N$ cyclic local (R-)symmetries, 

1105: the authors of \cite{Chemtob} have shown that some residual symmetries of the three types, GLP, GBP or 

1106: GMP, are possible within the NMSSM.

1107: 

1108: 

1109: {\bf $\bullet$ Mass matrix:}

1110: In this new framework, the Lagrangian containing the neutralino masses is the identical as (\ref{LAGmass})

1111: but with a different \rpv part of the mass matrix mixing neutrinos and neutralinos:

1112: \bea

1113: \xi^\prime_{\rpv} = 

1114: \left( 

1115: \begin{array}{ccccc}

1116: 0 & 0 & 0 & \lambda_1 \langle s \rangle  & \lambda_1 v_u  \\

1117: 0 & 0 & 0 & \lambda_2 \langle s \rangle  & \lambda_2 v_u   \\

1118: 0 & 0 & 0 & \lambda_3 \langle s \rangle  & \lambda_3 v_u   

1119: \end{array}

1120: \right).

1121: \label{RPVmassBIS}

1122: \eea

1123: Note the presence of the new mixings between $\tilde s$ and $\nu_i$. 

1124: 

1125: 

1126: \subsection{Effective neutrino mass}

1127: 

1128: {\bf $\bullet$ Mass expression:}

1129: Since we restrict to the situation $|\lambda_i/\lambda|<10^{-1}$, the effective neutrino mass matrix 

1130: is still given in a good approximation by the see-saw formula:

1131: \begin{equation}

1132: m_{\nu} = - \xi^\prime_{\rpv} \ {\cal M}_{NMSSM}^{-1} \ \xi_{\rpv}^{\prime \ T}.

1133: \label{seesawBIS}

1134: \end{equation}

1135: From Eq.(\ref{NMSSMmass}), Eq.(\ref{RPVmassBIS}) and Eq.(\ref{seesawBIS}), we derive analytically 

1136: the following effective Majorana neutrino mass matrix,

1137: \begin{equation}

1138: m_{\nu i j} = \frac{\lambda_i}{\lambda} \frac{\lambda_j}{\lambda} 

1139: \frac{M_1 M_2 (2 \kappa \mu / \lambda)}{Det({\cal M}_{NMSSM})}

1140: \bigg (

1141: \frac{\lambda^2 v_u v_d}{2 \kappa \mu / \lambda} + \frac{\mu}{2}

1142: \bigg )

1143: 2 \mu

1144: M_Z^2 [ \frac{\sin^2\theta_W}{M_1} + \frac{\cos^2\theta_W}{M_2} ] \cos^2 \beta.

1145: \label{MnuEFFBIS}

1146: \end{equation}

1147: 

1148: 

1149: In this scenario, we remark in Eq.(\ref{MnuEFFBIS}) that the specific 

1150: ratio particularly relevant for the discussion becomes $\lambda_i/\lambda$ instead of $\mu_i/\mu$

1151: (as in scenario I), 

1152: since one has here (in terms of effective $\mu_i$ and $\mu$ parameters): 

1153: $$

1154: \frac{\mu_i}{\mu}=\frac{\lambda_i \langle s \rangle }{\lambda \langle s \rangle }=\frac{\lambda_i}{\lambda}.

1155: $$

1156: 

1157: 

1158: {\bf $\bullet$ Origin of smallness:}

1159: Once again, we see on the neutrino mass matrix (\ref{MnuEFFBIS}) that there is a possible

1160: source of suppression from an approximate cancellation between the two terms in brackets. 

1161: This neutrino mass suppression has a different source from the other suppression issued from 

1162: the smallness of ratio $|\lambda_i/\lambda|$ (see Eq.(\ref{MnuEFFBIS})). The smallness of 

1163: this ratio would provide an interpretation of the neutrino mass hierarchy problem by introducing 

1164: another hierarchy, namely the hierarchy between the fundamental parameters $\lambda_i$ and 

1165: $\lambda$. 

1166: 

1167: 

1168: \begin{figure}[t]

1169: \begin{center}\unitlength1mm

1170: \SetScale{2.8}

1171: \begin{boldmath}

1172: %

1173: \begin{picture}(80,80)(0,-10)

1174: \Line(0,0)(15,0)\Line(65,0)(15,0)

1175: %\ArrowLine(80,0)(65,0)

1176: \Line(78,0)(65,0)

1177: \DashLine(25,0)(25,30){2}

1178: \Text(25,-4.5)[c]{$\lambda_i v_{u}$}

1179: \DashLine(55,0)(55,30){2}

1180: \Text(55,-4.5)[c]{$\lambda_j v_{u}$}

1181: \Text(52.6,29.6)[l]{$\times$}

1182: \Text(55,35)[c]{$\langle {h}^0_{u} \rangle $}

1183: \Text(22.9,29.6)[l]{$\times$}

1184: \Text(25,35)[c]{$\langle {h}^0_{u} \rangle $}

1185: \Text(73,0)[r]{$<$}

1186: %\Text(61,0)[r]{$>$}

1187: \Text(-1,0)[r]{$\nu_i$}

1188: \Text(8,0)[r]{$>$}

1189: %\Text(22,0)[r]{$<$}

1190: \Text(80,0)[l]{$\nu_j$}

1191: \Text(40,-6)[c]{$\widetilde{s}$}

1192: \Text(33,0)[r]{$<$}

1193: \Text(50,0)[r]{$>$}

1194: \Text(40,5)[c]{$2\kappa \mu/\lambda$}

1195: \Text(40,0)[c]{$\times$}

1196: %\Text(15,0)[c]{x}

1197: %\Text(65,0)[c]{x}

1198: %\Text(15,5)[c]{$\mu_i$}

1199: %\Text(65,5)[c]{$\mu_j$}

1200: \end{picture}

1201: \end{boldmath}

1202: \caption{Feynman diagram for the contribution to Majorana neutrino mass which arises in the 

1203: NMSSM through the trilinear coupling of Eq.(\ref{WmII}).}

1204: \protect\label{fig:diagramNMSSMBIS}

1205: \end{center}

1206: \end{figure}

1207: 

1208: 

1209: This possible cancellation in Eq.(\ref{MnuEFFBIS}) can be understood from a diagramatic point of view,

1210: as before. Indeed, in the present framework, the Majorana neutrino mass still receives contributions 

1211: from the previous exchanges of gauginos and singlino represented in diagrams (\ref{fig:diagramMSSM}), 

1212: except that the $\mu_i$ mass insertions in these diagrams come 

1213: now through the VEV of $s$ and should be parameterized instead by $\lambda_i \langle s \rangle$. 

1214: Furthermore, there is 

1215: new possible exchange of singlino which we have drawn in Fig.(\ref{fig:diagramNMSSMBIS}). This contribution

1216: is due to the new trilinear \rpv term of superpotential (\ref{WmII}). The above approximate cancellation 

1217: between the two terms in brackets entering neutrino mass (\ref{MnuEFFBIS}) would originate from a 

1218: compensation between the two types of process contributing to neutrino mass: the exchange of gauginos 

1219: (Fig.(\ref{fig:diagramMSSM})(a)) and the exchanges of a singlino (Fig.(\ref{fig:diagramMSSM})(b) and 

1220: Fig.(\ref{fig:diagramNMSSMBIS})).

1221: 

1222: 

1223: This cancellation-like source of neutrino mass suppression requires some fine tuning on NMSSM parameters. 

1224: It is, once more, $\lambda$ that faces the strongest fine tuning. Nevertheless, this fine tuning on $\lambda$ 

1225: decreases greatly as $| \mu |$ and $| \kappa |$ ($\lambda$ and $\tan \beta$) get smaller (larger).

1226: 

1227: \subsection{Numerical results and discussion}

1228: 

1229: \begin{table}[t]

1230: \begin{center}

1231: \begin{tabular}{c|c|c|c|c|c|c|c}

1232:  & $\lambda$ & $\mu$ & $\tan \beta$ & $M_1$ & $M_2$ & $\lambda_1$ & $- \kappa$  

1233: \\

1234: &  & $\mbox{[GeV]}$ &  & $\mbox{[GeV]}$ & $\mbox{[GeV]}$ & &

1235: \\

1236: \hline

1237: A & 0.7 &  110 &  50   & 100 & 100 & $7 \ 10^{-2}$ & $(1.71783 - 1.7177) \ 10^{-2}$ 

1238: \\

1239: &  &   &    &  &  & $7 \ 10^{-4}$ & $(16.2 - 7) \ 10^{-3}$ 

1240: \\

1241: \hline

1242: B & 0.7 &  300 &  50   & 50 & 500 & $7 \ 10^{-2}$ & $(2.30954 - 2.3095) \ 10^{-3}$ 

1243: \\

1244: &  &   &   & & & $7 \ 10^{-4}$ & $(2.26 - 1.8) \ 10^{-3}$ 

1245: \\

1246: \hline

1247: C & 0.4 &  -110 &  30   & 100 & 100 & $4 \ 10^{-2}$ & $(5.33839 - 5.3385) \ 10^{-3}$ 

1248: \\

1249: & &  &   &  &  & $4 \ 10^{-4}$ & $(5.41 - 6) \ 10^{-3}$ 

1250: \end{tabular}

1251: \caption{Sets (A,B,C) of values, for the parameters entering the whole neutralino mass matrix, 

1252: which reproduce the correct neutrino mass. The two values of parameter $\kappa$ correspond to 

1253: the neutrino masses $m_{\nu}= 0.1 \mbox{eV} - 1 \mbox{eV}$ ($m_{\nu}$ being defined 

1254: via Eq.(\ref{MnuEFFBIS})), respectively. As the one-flavor case is considered here, 

1255: the flavor index $i$ of \rpv coupling constant $\lambda_i$ takes only the value $i=1$.}

1256: \label{tab:parII}

1257: \end{center}

1258: \end{table}

1259: 

1260: 

1261: {\bf $\bullet$ Neutrino mass suppression:}

1262: In Table \ref{tab:parII}, we show characteristic sets of parameters for which the neutrino mass 

1263: (\ref{MnuEFFBIS}) at one flavor is equal to $0.1 \mbox{eV}$ and $1 \mbox{eV}$, covering the 

1264: range of values motivated by experimental data (Eq.(\ref{bdTOT})).

1265: 

1266: In Table \ref{tab:parII}, the $\kappa$ value is fixed by the other parameters via formula (\ref{MnuEFFBIS}). 

1267: Here, we have chosen to fix $\kappa$ as it is direct to solve Eq.((\ref{MnuEFFBIS})) in term of this parameter. 

1268: 

1269: Let us comment on the results in Table \ref{tab:parII}. $\kappa$ is chosen negative so that 

1270: the cancellation between the terms in brackets of expression (\ref{MnuEFFBIS}) is effective. 

1271: This cancellation can be only partially responsible for the neutrino mass reduction, as for 

1272: points in the table with $| \lambda_1/\lambda | = 10^{-3}$ ($\lambda_1 = 4,7 \ 10^{-4}$). 

1273: This cancellation can also be the principal mechanism that suppresses the neutrino mass,

1274: as for the points with $| \lambda_1/\lambda | = 10^{-1}$ ($\lambda_1 = 4,7 \ 10^{-2}$).

1275: Then the wanted neutrino mass suppression is reached without requiring any highly 

1276: hierarchical pattern.

1277: 

1278: 

1279: \begin{table}[t]

1280: \begin{center}

1281: \begin{tabular}{c|c|c|c}

1282: & $m_{\tilde \chi^0_1}$  & $m_{\tilde \chi^0_5}$  &  ${\cal F}_\lambda$  

1283: \\

1284: & $\mbox{[GeV]}$ & $\mbox{[GeV]}$ &

1285: \\

1286: \hline

1287: A & 34 & 193 &   $(0.2 - 1.9) \ 10^{-5}$ 

1288: \\

1289: & 33 & 193 & $(2.1 - 196) \ 10^{-2}$   

1290: \\

1291: \hline

1292: B & 7 & 520 &  $(0.7 - 6.9) \ 10^{-6}$ 

1293: \\

1294: & 7 & 520  & $(0.7 - 9.4) \ 10^{-2}$   

1295: \\

1296: \hline

1297: C & 18 & 180  &  $(0.4 - 4.1) \ 10^{-6}$ 

1298: \\

1299: & 18 & 180 & $(0.4 - 3.3) \ 10^{-2}$ 

1300: \end{tabular}

1301: \caption{Lowest [$m_{\tilde \chi_1^0}$] and highest [$m_{\tilde \chi_5^0}$] neutralino masses  

1302: for the points A,B,C of parameter space presented in Table \ref{tab:parII}. Together with these 

1303: masses, we also give the value of fine tuning quantity ${\cal F}_\lambda$ defined in text for the 

1304: $\lambda$ parameter. The two values of ${\cal F}_\lambda$ correspond respectively to the two 

1305: $\kappa$ values in Table \ref{tab:parII} (leading to $m_{\nu}= 0.1 \mbox{eV} - 1 \mbox{eV}$). 

1306: For each point, the first and second lines are respectively associated to $\lambda_1=4,7 \ 10^{-2}$ 

1307: and $\lambda_1=4,7 \ 10^{-4}$, as in Table \ref{tab:parII}.}

1308: \label{tab:FTII}

1309: \end{center}

1310: \end{table}

1311: 

1312: 

1313: 

1314: {\bf $\bullet$ Fine tuning:}

1315: We quantify the fine tuning on $\lambda$ with variable (\ref{FTratio}), now defined with

1316: the neutrino mass (\ref{MnuEFFBIS}). On Table \ref{tab:FTII}, we give the values of this variable 

1317: ${\cal F}_\lambda$ for the points of parameter space shown in Table \ref{tab:parII} which reproduce 

1318: the correct neutrino masses through the compensation mechanism. A comparison of points A and B in 

1319: Table \ref{tab:FTII} shows that the fine tuning on $\lambda$ is softer if $| \mu |$ decreases. 

1320: In the same way, by comparing parameters A and C, one observes that the fine tuning is significantly 

1321: improved for larger values of $\lambda$ or $\tan \beta$. Table \ref{tab:FTII} also exhibits that the 

1322: fine tuning is soften for higher neutrino masses and smaller $| \lambda_1/\lambda |$ ratios. 

1323: 

1324: 

1325: 

1326: {\bf $\bullet$ Tachyons:}

1327: Unfortunately, it turns out that for any domain of the parameter space $\{ \tan \beta, \mu, M_1, M_2 \}$, 

1328: the fact of requiring the neutrino mass (\ref{MnuEFFBIS}) to be suppressed down to the $\mbox{eV}$ scale,

1329: at least partially through our cancellation mechanism (namely for $| \lambda_i/\lambda | \gtrsim 10^{-6}$), 

1330: imposes the ratio $| \kappa / \lambda |$ to be small, leading to the occurrence of unacceptable tachyons 

1331: in the CP-even Higgs sector. 

1332: 

1333: 

1334: We have checked this feature of our scenario by using the code NMHDECAY \cite{nmhdecay}

1335: which applies on the pure NMSSM 

1336: parameter space. However, this procedure is believed to be consistent since the additional trilinear 

1337: \rpv interaction $\lambda_i S L_i H_u$ in the superpotential (see Eq.(\ref{WmII})) is not expected to 

1338: induce considerable modifications in the scalar potential of the NMSSM. As a matter of fact, we

1339: have systematically restricted ourselves to the case $| \lambda_i/\lambda | \leq 10^{-1}$ 

1340: ({\it c.f.} Eq.(\ref{WNMSSM})).

1341: 

1342: 

1343: In a situation where the $| \lambda_i/\lambda |$ ratio would be of order unity or even larger, 

1344: giving rise to important changes in the NMSSM potential, it could happen that our cancellation 

1345: mechanism for neutrino mass suppression would be active without implying necessarily the 

1346: appearance of tachyons in the theory.

1347: 

1348: 

1349: Another way out of this theoretical problem is to focus on the particular case $A_\kappa = 0$,

1350: in which no tachyons emerge from the CP-even sector. This possibility is conceivable as the 

1351: trilinear soft supersymmetry breaking parameter $A_\kappa$, which was previously introduced in 

1352: Section \ref{NumResI}, does not affect the neutralino mass matrix (\ref{CHImass}) on which is 

1353: based our analysis. 

1354: 

1355: 

1356: \begin{figure}[t]

1357: \unitlength 1mm

1358: \SetScale{2.8}

1359: \begin{boldmath}

1360: \begin{center}

1361: %

1362: \begin{picture}(60,30)(0,-10)

1363: \Line(0,0)(20,0)

1364: \Line(45,15)(20,0)

1365: \DashLine(45,-15)(20,0){2}

1366: %\ArrowLine(0,0)(15,0)

1367: %\DashLine(15,0)(15,15){2}

1368: %\DashLine(45,0)(45,15){2}

1369: %\Line(45,0)(15,0)

1370: %\DashLine(45,0)(45,10)

1371: %\ArrowLine(60,0)(45,0)

1372: \Text(-2,0)[r]{$\widetilde s$}

1373: \Text(22,-4)[r]{$\lambda_i$}

1374: \Text(45,17)[l]{$\ell_i$}

1375: \Text(45,-18)[c]{$h^+_{u}$}

1376: \Text(30,-19)[c]{(a)}

1377: %\Text(30,5)[c]{$m_{\chi_{\alpha}}$}

1378: %\Text(15,15)[c]{X}

1379: %\Text(15,0)[c]{x}

1380: %\Text(45,15)[c]{X}

1381: %\Text(15,5)[c]{$\mu_i$}

1382: %\Text(45,5)[c]{$\mu_j$}

1383: \end{picture}

1384: %%

1385: \hspace{1cm}\begin{picture}(60,30)(0,-10)

1386: \Line(0,0)(20,0)

1387: \Line(45,15)(20,0)

1388: \DashLine(45,-15)(20,0){2}

1389: \Line(45,15)(45,-15)

1390: \DashLine(45,15)(65,15){2}

1391: \Line(45,-15)(65,-15)

1392: %\ArrowLine(0,0)(15,0)

1393: %\DashLine(15,0)(15,15){2}

1394: %\DashLine(45,0)(45,15){2}

1395: %\Line(45,0)(15,0)

1396: %\DashLine(45,0)(45,10)

1397: %\ArrowLine(60,0)(45,0)

1398: \Text(-2,0)[r]{$\widetilde s$}

1399: \Text(22,-4)[r]{$\lambda_j$}

1400: \Text(35,13.5)[l]{$\ell_j$}

1401: \Text(35,-13)[c]{$h^+_{u}$}

1402: \Text(45,-18)[c]{$\lambda_i$}

1403: \Text(45,18)[c]{$\lambda_j$}

1404: \Text(30,-19)[c]{(b)}

1405: %\Text(30,5)[c]{$m_{\chi_{\alpha}}$}

1406: \Text(47,0)[c]{$\widetilde s$}

1407: %\Text(15,0)[c]{x}

1408: \Text(65,17)[c]{$h^+_{u}$}

1409: %\Text(15,5)[c]{$\mu_i$}

1410: \Text(65,-17)[c]{$\ell_i$}

1411: \end{picture}

1412: %\begin{picture}(60,30)(0,-10)

1413: %\Line(0,0)(20,0)

1414: %\Line(45,15)(20,0)

1415: %\DashLine(45,-15)(20,0){2}

1416: %\ArrowLine(0,0)(15,0)

1417: %\DashLine(15,0)(15,15){2}

1418: %\DashLine(45,0)(45,15){2}

1419: %\Line(45,0)(15,0)

1420: %\DashLine(45,0)(45,10)

1421: %\ArrowLine(60,0)(45,0)

1422: %\Text(-2,0)[r]{$S$}

1423: %\Text(45,17)[l]{$\bar \ell_j$}

1424: %\Text(45,-18)[c]{$\bar{h_u}$}

1425: %\Text(22,-4)[r]{$\lambda_j$}

1426: %\Text(30,5)[c]{$m_{\chi_{\alpha}}$}

1427: %\Text(15,15)[c]{X}

1428: %\Text(15,0)[c]{x}

1429: %\Text(45,15)[c]{X}

1430: %\Text(15,5)[c]{$\mu_i$}

1431: %\Text(45,5)[c]{$\mu_j$}

1432: %\end{picture}

1433: %%

1434: \end{center}

1435: \end{boldmath}\vskip 0.3cm

1436: \caption{CP-asymmetry decay diagrams at tree (a) and loop (b) level in the NMSSM with

1437: \rpv trilinear couplings.}

1438: \protect\label{lepto}

1439: \end{figure}

1440: 

1441: 

1442: {\bf $\bullet$ Leptogenesis:}

1443: Anyway, within this second scenario, the suppression of neutrino mass scale can 

1444: be insured by a small $| \lambda_i/\lambda |$ value, so that the tachyonic regions associated to the 

1445: cancellation mechanism are avoided. Then this scenario still possesses a new interesting phenomenological 

1446: feature: the extra singlet of the NMSSM can produce, via decay channels involving \rpv trilinear couplings, 

1447: a thermal leptogenesis. This leptogenesis can be converted into the baryonic sector through sphaleron induced 

1448: processes, explaining then the baryon asymmetry of the universe. The lepton asymmetry arises through the 

1449: out-of-equilibrium decay of the singlet in a L and CP-violating way, according to Sakharov's constraints  

1450: \cite{Sakharov}.

1451: Indeed, the CP-asymmetry may be generated from the interference between the tree level diagram of Fig.(\ref{lepto})(a)

1452: and one-loop diagrams such as the one drawn in Fig.(\ref{lepto})(b). Only the \rpv trilinear couplings

1453: of Eq.(\ref{WmII}) enter the two diagrams in Fig.(\ref{lepto}). In that case, the CP-asymmetry

1454: \bea

1455: \epsilon={\Gamma(S\to \ell H)-\Gamma(S\to \bar \ell H)\over \Gamma(S\to \ell H)+\Gamma(S\to \bar \ell  H)}

1456: \eea

1457: would be proportional to $\sum_j \lambda_i^2 \lambda_j^2 f$, 

1458: where $f$ is the loop 

1459: function. There exist other types of diagrams, generating a CP-asymmetry, which involve the lepton Yukawa

1460: ($Y^\ell_{ij}$) and \rpv trilinear ($\lambda_i$) coupling constants.

1461: 

1462: 

1463: \section{Conclusion}

1464: \label{conclu}

1465: 

1466: 

1467: First, we have considered the NMSSM, which solves the $\mu$ problem, in the presence of bilinear 

1468: \rpv interactions $\mu_i L_i H_u$ (scenario I). 

1469: In this context, we have found that a cancellation mechanism 

1470: arises for suppressing the Majorana neutrino mass and thus provides an interpretation to the smallness

1471: of neutrino mass compared to the electroweak scale. This mechanism, which relies on the existence of 

1472: the gauge-singlet $S$ introduced by the NMSSM, offers a solution for the neutrino mass problem 

1473: which is interestingly connected to the solution for the $\mu$ problem.

1474: \\

1475: More precisely, by using the NMHDECAY program, we have obtained various characteristic points of 

1476: the NMSSM parameter space which satisfy the experimental constraints from collider physics, fulfill the 

1477: theoretical consistency conditions (physical minimum, no Landau singularity,\dots) and simultaneously

1478: {\it generate} neutrino masses of order of the $\mbox{eV}$ scale through our cancellation mechanism. 

1479: By the verb `generate', we mean here that small neutrino mass values are effectively produced 

1480: without introducing a strong hierarchy between the fundamental parameters. Indeed, in the basis where

1481: $\langle  \tilde \nu_i  \rangle =0$, the obtained parameters lead to neutrino masses $m_\nu \in [0.1,1] \mbox{eV}$ 

1482: with $10^{-3} \lesssim |\mu_i / \mu| \lesssim 10^{-1}$ (the extreme values given here, for the ranges of 

1483: neutrino mass and $|\mu_i / \mu|$ ratio, are not corresponding to each other). 

1484: \\

1485: In comparison, the see-saw mechanism suppresses sufficiently neutrino masses by 

1486: introducing an high hierarchy between the Dirac and Majorana masses. Furthermore, in the

1487: MSSM with a non-vanishing $\mu_i L_i H_u$ term, realistic neutrino masses are achieved for

1488: $|\mu_i / \mu| \sim 10^{-7}$ typically. Finally, in the new version of the MSSM suggested recently 

1489: in \cite{AltII}, which constitutes an alternative to our scenario as it addresses both the $\mu$ 

1490: value and neutrino mass problems, a stronger hierarchy of $\sim 10^{-6}$ is required between the 

1491: the neutrino and top quark Yukawa coupling constants. 

1492: \\

1493: Nevertheless, our new cancellation mechanism for neutrino mass suppression needs a certain fine tuning 

1494: on some NMSSM parameters. For some of the obtained parameters mentioned above, that generate neutrino

1495: masses around the $\mbox{eV}$, the most important fine tuning reaches the acceptable level of 

1496: $\sim 3 \ 10^{-1}$ ($\sim 10^{-3}$) for $|\mu_i / \mu| \simeq 10^{-3}$ ($\simeq 10^{-1}$).

1497: \\ 

1498: The continuation of this study \cite{prep} would be the combination 

1499: of tree and one-loop contributions with three flavors 

1500: in order to accommodate all the last data on neutrino masses and leptonic mixing angles.

1501: 

1502: Secondly, we have studied another attractive version of this model (scenario II), namely the NMSSM with 

1503: \rpv $\mu_i$-like interactions generated naturally by the VEV of the $S$ scalar component, through 

1504: the trilinear term $\lambda_i S L_i H_u$. There the same kind of cancellation mechanism can occur for

1505: the neutrino mass suppression. Based on this mechanism, we have easily found parameters which give rise 

1506: to $m_\nu \in [0.1,1] \mbox{eV}$ for $10^{-3} \leq |\lambda_i / \lambda| \leq 10^{-1}$ corresponding

1507: to a quite soft hierarchy. The associated fine tuning can reach $\sim 1$ ($\sim 10^{-5}$) for 

1508: $|\lambda_i / \lambda| = 10^{-3}$ ($= 10^{-1}$). However, here, the cancellation mechanism seems to 

1509: imply the occurrence of tachyons in the CP-even sector, at least in the simplest form of the NMSSM.

1510: So one should think of some way out, like restricting to the particular situation where $A_\kappa$ 

1511: vanishes. 

1512: 

1513: 

1514: It would also be interesting to find an independent theoretical reason for the compensation, between 

1515: the two types of process exchanging the gauginos and singlino, which explains this new cancellation 

1516: mechanism of neutrino mass suppression. In the same philosophy as for the see-saw mechanism, where

1517: the Dirac/Majorana mass hierarchy introduced finds a natural realization within the framework of the 

1518: $SO(10)$ GUT model.

1519: 

1520: 

1521: Let us finish by commenting on the specific and rich phenomenology of the NMSSM with additional 

1522: \rpv $\mu_i$-like interactions. We have discussed the fact that such a framework opens the possibility 

1523: of new leptogenesis scenarios. This framework also leads to new decay channels for the Lightest 

1524: Supersymmetric Particle (LSP). For instance, in the case where the LSP is the lightest neutralino,

1525: it can decays as $\tilde \chi^0_1 \to \nu_i Z^0$ and $\tilde \chi^0_1 \to l_j^\pm W^\mp$ via the 

1526: \rpv $\mu_i$-like mixings $\tilde h^0_u \nu_i$ or $\tilde s \nu_i$. The value of the LSP life time associated to 

1527: these new decays

1528: \footnote{In our scenario, the $|\mu_i/\mu|$ ratio can reach values close to unity which tends to

1529: increase significantly the width of these new LSP decays, except if the $\tilde \chi^0_1$ is mainly 

1530: composed by $\tilde B^0$, $\tilde W^0_3$ and/or $\tilde h^0_d$.}

1531: is fundamental in regard of the collider physics (if the LSP decays inside the

1532: detectors, the typical supersymmetric signatures are multi-jets/leptons instead of missing energy) 

1533: as well as of the dark matter problem (the LSP remains a good WIMP candidate only if it is stable,  

1534: relatively to the age of the universe).

1535: 

1536: 

1537: 

1538: \section*{Acknowledgments}

1539: 

1540: The authors are grateful to A.~Djouadi for stimulating discussions and 

1541: G.~Bhattacharyya for reading 

1542: the manuscript. It is also a pleasure to thank M.~Chemtob, U.~Ellwanger, P.~Fayet, 

1543: C.~Hugonie, C.~Mun\~oz and C.~A.~Savoy for fruitful conversations.

1544: 

1545: 

1546: \newpage

1547: 

1548: \begin{thebibliography}{99} 

1549: \bibitem{Susskind} L.~Susskind, \Journal{\PR}{104}{1881}{1984}.

1550: \bibitem{mu} J.~E.~Kim and H.~P.~Nilles, \Journal{\PLB}{138}{150}{1984}.

1551: \bibitem{NMSSM} 

1552: H.~P.~Nilles {\it et al.}, \Journal{\PLB}{120}{346}{1983};

1553: \\ J.~M.~Fr\`ere {\it et al.}, \Journal{\NPB}{222}{11}{1983};

1554: \\ J.~P.~Derendinger and C.~A.~Savoy, \Journal{\NPB}{237}{307}{1984}; 

1555: \\ J.~R.~Ellis {\it et al.}, \Journal{\PRD}{39}{844}{1989};

1556: \\ M.~Drees, {\it Int. J. Mod. Phys}. A {\bf 4}, 3635 (1989);

1557: \\ U.~Ellwanger {\it et al.}, \Journal{\PLB}{315}{331}{1993};

1558: \\ S.~F.~King and P.~L.~White, \Journal{\PRD}{52}{4183}{1995};

1559: \\ F.~Franke and H.~Fraas, {\it Int. J. Mod. Phys.} A {\bf 12}, 479 (1997).

1560: \bibitem{NMpheno} 

1561: U.~Ellwanger {\it et al.}, {\it Z. Phys.} C {\bf 67}, 665 (1995); 

1562: \\ U.~Ellwanger {\it et al.}, \Journal{\NPB}{492}{21}{1997};

1563: \\ U.~Ellwanger and C.~Hugonie, {\it Eur. Phys. J.} C {\bf 13}, 681 (2000); 

1564: \\ A.~Dedes {\it et al.}, \Journal{\PRD}{63}{055009}{2001};

1565: \\ U.~Ellwanger {\it et al.}, {\tt arxiv: hep-ph/0111179};

1566: \\ U.~Ellwanger {\it et al.}, \Journal{\JHEP}{0507}{041}{2005};

1567: \\ U.~Ellwanger and C.~Hugonie, \Journal {\PLB}{623}{93}{2005};

1568: \\ G.~Belanger {\it et al.}, \Journal{JCAP}{0509}{001}{2005}.

1569: \bibitem{littleFT} M.~Bastero-Gil {\it et al.}, \Journal{\PLB}{489}{359}{2000}.

1570: \bibitem{operators} 

1571: S.~A.~Abel, \Journal{\NPB}{480}{55}{1996};

1572: \\ C.~Panagiotakopoulos and K.~Tamvakis, \Journal{\PLB}{446}{224}{1999}.

1573: \bibitem{wall} S.~A.~Abel {\it et al.}, \Journal{\NPB}{454}{663}{1995}.

1574: \bibitem{rpar1} 

1575: G.~Farrar and P.~Fayet, \Journal{\PLB}{76}{575}{1978}; 

1576: \\ S.~Weinberg, \Journal{\PRD}{26}{287}{1982}; 

1577: \\ N.~Sakai and T.~Yanagida, \Journal{\NPB}{197}{533}{1982}; 

1578: \\ C.~Aulakh and R.~Mohapatra, \Journal{\PLB}{119}{136}{1982}.

1579: \bibitem{rpar2} 

1580: F.~M.~Borzumati, Y.~Grossman, E.~Nardi and Y.~Nir, \Journal{\PLB}{384}{123}{1996};  

1581: \\ E.~Nardi, \Journal{\PRD}{55}{5772}{1997};  

1582: \\ H.~P.~Nilles and N.~Polonsky, \Journal{\NPB}{499}{33}{1997}.

1583: \bibitem{reviewsD} 

1584: H.~K.~Dreiner, {\it ``Perspectives on Supersymmetry''}, 

1585: Edited by G.~L.~Kane, World Scientific, {\tt arxiv: hep-ph/9707435}; 

1586: \\ B.~Allanach, A.~Dedes and H.~K.~Dreiner, \Journal{\PRD}{60}{075014}{1999}.

1587: \bibitem{reviewsB}

1588: G.~Bhattacharyya {\it et al.}, {\it Mod. Phys. Lett.} A {10}, 1583 (1995);

1589: \\ G.~Bhattacharyya and D.~Choudhury, {\it Mod. Phys. Lett.} A {10}, 1699 (1995);

1590: \\ G.~Bhattacharyya, {\it Nucl. Phys. Proc. Suppl.} A {\bf 52}, 83 (1997), 

1591: {\tt arxiv: hep-ph/9608415};

1592: \\ G.~Bhattacharyya, Invited talk presented at `Beyond the Desert', 

1593: Castle Ringberg, Tegernsee, Germany, 8-14 June 1997, {\tt arxiv: hep-ph/9709395}; 

1594: \\ G.~Bhattacharyya and A.~Raychaudhuri, \Journal{\PRD}{57}{3837}{1998}.

1595: \bibitem{HalfWay} R.~Barbier {\it et al.}, {\tt arxiv: hep-ph/9810232}.

1596: \bibitem{Marc} M.~Chemtob, {\it Prog. Part. Nucl. Phys.} {\bf 54}, 71 (2005).

1597: \bibitem{PhysRep} R.~Barbier {\it et al.}, \Journal{\PR}{420}{1}{2005}.

1598: \bibitem{Chemtob} M.~Chemtob and P.~N.~Pandita, \Journal{\PRD}{73}{055012}{2006}.

1599: \bibitem{Ibanez} L.~E.~Ib\'a\~nez and G.~G.~Ross, \Journal{\PLB}{260}{291}{1991}; 

1600: \Journal{\NPB}{368}{3}{1992};

1601: \\ L.~E.~Ib\'a\~nez, \Journal{\NPB}{218}{514}{1982}.

1602: \bibitem{AlignI} L.~Hall and M.~Suzuki, \Journal{\NPB}{231}{419}{1984}.

1603: \bibitem{GUT1} G.~F.~Giudice and R.~Rattazzi, \Journal{\PLB}{406}{321}{1997}.

1604: \bibitem{GUT2} D.~Brahm amd L.~Hall, \Journal{\PRD}{40}{2449}{1989}.

1605: \bibitem{GUT3} K.~Tamvakis, \Journal{\PLB}{382}{251}{1996}; \Journal{\PLB}{307}{251}{1996}.

1606: \bibitem{GUT4} A.~Y.~Smirnov and F.~Vissani, \Journal{\NPB}{460}{37}{1996}.

1607: \bibitem{GUT5} R.~Barbieri {\it et al.}, \Journal{\PLB}{407}{250}{1997}.

1608: \bibitem{string} 

1609: M.~Bento {\it et al.}, \Journal{\NPB}{292}{400}{1987};

1610: \\ N.~Ganoulis {\it et al.}, \Journal{\NPB}{323}{374}{1989}.

1611: \bibitem{MixNeut} 

1612: C.~S.~Aulakh and R.~N.~Mohapatra, \Journal{\PLB}{121}{14}{1983};

1613: \\ J.~R.~Ellis {\it et al.}, \Journal{\PLB}{150}{142}{1985};

1614: \\ G.~G.~Ross and J.~W.~F.~Valle, \Journal{\PLB}{151}{375}{1985};

1615: \bibitem{AltI} 

1616: A.~Masiero and J.~W.~F.~Valle, \Journal{\PLB}{251}{273}{1990};

1617: \\ J.~Rom\~ao {\it et al.}, \Journal{\PLB}{292}{329}{1992};

1618: \\ M.~Hirsch {\it et al.}, \Journal{\PRD}{70}{073012}{2004}.

1619: \bibitem{AltII} D.~E.~Lopez-Fogliani and C.~Mu\~noz, {\tt arxiv: hep-ph/0508297}.

1620: \bibitem{nmhdecay}

1621: U.~Ellwanger, J.~F.~Gunion and C.~Hugonie, \Journal{\JHEP}{0502}{066}{2005};

1622: \\ U.~Ellwanger and C.~Hugonie, {\tt arxiv: hep-ph/0508022}.

1623: \bibitem{DreinerRoss} H.~K.~Dreiner and G.~G.~Ross, \Journal{\NPB}{365}{597}{1991}.

1624: \bibitem{LoopB} K.~S.~Babu and R.~N.~Mohapatra, \Journal{\PRL}{64}{1705}{1990}.

1625: \bibitem{MoDrei} H.~K.~Dreiner and G.~Moreau, \Journal{\PRD}{67}{055005}{2003}. 

1626: \bibitem{GroHab} 

1627: Y.~Grossman and H.~E.~Haber, \Journal{\PRL}{78}{3438}{1997};

1628: \\ Y.~Grossman and H.~E.~Haber, \Journal{\PRD}{59}{093008}{1999};

1629: \\ Y.~Grossman and H.~E.~Haber, \Journal{\PRD}{63}{075011}{2001}.

1630: \bibitem{Valle} M.~Maltoni, T.~Schwetz, M.~A.~Tortola and J.~W.~F.~Valle, 

1631: {\it New J. Phys.} {\bf 6}, 122 (2004), {\tt arxiv: hep-ph/0405172}. 

1632: \bibitem{MSSMrpvA} R.~Hempfling, \Journal{\NPB}{478}{3}{1996}.

1633: \bibitem{Kang} E.~J.~Chun and S.~K.~Kang, \Journal{\PRD}{61}{075012}{2000}. 

1634: \bibitem{Hi} M.~Hirsch {\it et al.}, \Journal{\PRD}{62}{113008}{2000}.

1635: \bibitem{DL1} S.~Davidson and M.~Losada, \Journal{\JHEP}{0005}{021}{2000}.

1636: \bibitem{AM1} A.~Abada and M.~Losada, \Journal{\NPB}{585}{45}{2000}.

1637: \bibitem{DL3} S.~Davidson, M.~Losada and N.~Rius, \Journal{\NPB}{587}{118}{2000}.

1638: \bibitem{AM2} A.~Abada and M.~Losada, \Journal{\PLB}{492}{310}{2000}. 

1639: \bibitem{adl} A.~Abada, S.~Davidson and M.~Losada, \Journal{\PRD}{65}{075010}{2002}.

1640: \bibitem{DL2} S.~Davidson and M.~Losada, \Journal{\PRD}{65}{075025}{2002}.

1641: \bibitem{abl} A.~Abada, G.~Bhattacharyya and M.~Losada, \Journal{\PRD}{66}{071701}{2002}, 

1642: Rapid Communication. 

1643: \bibitem{Borz} F.~Borzumati and J.~S.~Lee, \Journal{\PRD}{66}{115012}{2002}.

1644: \bibitem{Chun} E.~J.~Chun {\it et al.}, \Journal{\PLB}{557}{233}{2003}.

1645: \bibitem{MSSMrpvB} M.~A.~Diaz {\it et al.}, \Journal{\PRD}{68}{013009}{2003},

1646: Erratum-ibid. D {\bf 71}, 059904 (2005).

1647: \bibitem{cancel} Y.~Grossman and S.~Rakshit, \Journal{\PRD}{69}{093002}{2004}

1648: \bibitem{107} Y.~Farzan, O.~L.~G.~Peres and A.~Yu.~Smirnov, \Journal{\NPB}{612}{59}{2001}.

1649: \bibitem{108} J.~Bonn {\it et al.}, 

1650:    {\it Nucl. Phys. Proc. Suppl.} {\bf 91}, 273 (2001). 

1651: \bibitem{109} V.~M.~Lobashev {\it et al.}, 

1652:    \Journal{\PLB}{460}{227}{1999};

1653:    {\it Nucl. Phys. Proc. Suppl.} {\bf 77}, 327 (1999);

1654:                                   {\bf 91}, 280 (2001).

1655: \bibitem{cosmobound} S.~Hannestad, \Journal {JCAP}{0305}{004}{2003}.

1656: \bibitem{darkNMSSM} D.~G.~Cerdeno {\it et al.}, \Journal{\JHEP}{0412}{048}{2004};

1657: \bibitem{LEPchargino} LEP SUSY Working Group, LEPSUSYWG Note/02-04.1.

1658: \bibitem{PDG} Particle Data Group, S.~Eidelman {\it et al.}, \Journal{\PLB}{592}{1}{2004}.

1659: \bibitem{AlignII} T.~Banks, Y.~Grossman, E.~Nardi and Y.~Nir, \Journal{\PRD}{52}{5319}{1995}.

1660: \bibitem{davidson} S.~Davidson and J.~R.~Ellis, 

1661: \Journal{\PLB}{39}{210}{1997}; 

1662: \Journal{\PRD}{56}{4182}{1997}.

1663: \bibitem{Sakharov} A.~D.~Sakharov, {\it JETP Lett.} 5:24 (1967).

1664: \bibitem{prep} In preparation, in collaboration with G.~Bhattacharyya.

1665: \end{thebibliography} 

1666: 

1667: 

1668: 

1669: \end{document}

1670: \bye

1671: 

1672: 

1673: 

1674: 

1675: 

1676: 

1677: 

1678: