1: \documentclass[12pt]{article}

2: 

3: \usepackage{a4,amsmath,amssymb,hyperref}

4: \usepackage{graphicx}

5: 

6: \oddsidemargin 0mm

7: \textwidth 16.5cm

8: \topmargin -1cm

9: \allowdisplaybreaks[4]

10: \sloppy

11: 

12: \renewcommand{\title}[1]{\vspace{10mm}\noindent{\Large{\bf #1}}\vspace{8mm}}

13: \newcommand{\authors}[1]{\noindent{\large #1}\vspace{3mm}}

14: \newcommand{\address}[1]{{\itshape #1\vspace{2mm}}}

15: 

16: \makeatletter

17: \def\section{\@startsection{section}{1}{\z@}{-3.25ex plus -1ex minus

18:     -.2ex}{1.5ex plus .2ex}{\normalfont\large\bfseries}}

19: \def\subsection{\@startsection{subsection}{1}{\z@}{-3.25ex plus -1ex

20:     minus -.2ex}{1.5ex plus .2ex}{\normalfont\itshape}}

21: 

22: \renewenvironment{thebibliography}[1]

23:          {\section*{References}\frenchspacing\small

24:           \begin{list}{[\arabic{enumi}]}

25:          {\usecounter{enumi}\parsep=2pt\topsep 0pt

26:          \settowidth{\labelwidth}{[#1]}

27:          \leftmargin=\labelwidth\advance\leftmargin\labelsep

28:          \rightmargin=0pt\itemsep=0pt\sloppy}}{\end{list}}

29: 

30: \makeatother

31: 

32: \begin{document}

33: 

34: 

35: \begin{titlepage}

36: 

37: \hspace*{\fill}%arXiv:YYMM.NNNN  

38: 

39: \begin{center}

40: 

41: \title{8D-spectral triple on 4D-Moyal space and\\[2mm]

42:   the vacuum of noncommutative gauge theory} 

43: 

44: \authors{Harald {\sc Grosse}$^1$ and Raimar {\sc Wulkenhaar}$^2$}

45: 

46: 

47: \address{$^{1}$\,Institut f\"ur Theoretische Physik, Universit\"at Wien\\

48: Boltzmanngasse 5, A-1090 Wien, Austria}

49: 

50: \address{$^{2}$\,Mathematisches Institut der Westf\"alischen

51:   Wilhelms-Universit\"at\\

52: Einsteinstra\ss{}e 62, D-48149 M\"unster, Germany}

53: 

54: \footnotetext[1]{harald.grosse@univie.ac.at}

55: \footnotetext[2]{raimar@math.uni-muenster.de}

56: 

57: 

58: \vskip 3cm

59: 

60: \textbf{Abstract} \vskip 3mm 

61: 

62: \begin{minipage}{14cm}%

63:   Observing that the Hamiltonian of the renormalisable scalar field

64:   theory on 4-dimensional Moyal space $\mathcal{A}$ is the square of a

65:   Dirac operator $\mathcal{D}$ of spectral dimension 8, we complete

66:   $(\mathcal{A},\mathcal{D})$ to a compact 8-dimensional spectral

67:   triple (with violated orientability axiom). We add another

68:   Connes-Lott copy and compute the spectral action of the

69:   corresponding $U(1)$-Yang-Mills-Higgs model. We find that in the

70:   Higgs potential the square $\phi^2$ of the Higgs field is shifted to

71:   $\phi\star \phi + \mathrm {const} \cdot X_\mu \star X^\mu$, where

72:   $X_\mu$ is the covariant coordinate. The classical field equations

73:   of our model imply that the vacuum is no longer

74:   given by a constant Higgs field, but both the Higgs and gauge fields

75:   receive non-constant vacuum expectation values. For pure Yang-Mills

76:   theory we compute the vacuum solution for the order parameter $X_\mu

77:   \star X^\mu$ explicitly in terms of modified Bessel and Struve

78:   functions.

79: \end{minipage}

80: 

81: 

82: \end{center}

83: 

84: \end{titlepage}

85: 

86: 

87: \section{Introduction}

88: 

89: \subsection{Renormalisable field theories on Moyal space}

90: 

91: Renormalisable field theories on Moyal space are by now in mature

92: state. In the first renormalisation proof \cite{Grosse:2004yu}, the

93: matrix base of the Moyal plane was a central philosophy, because we

94: wanted to avoid convergence subtleties with the oscillating integrals

95: in momentum space. We traded the simple matrix product interaction in

96: for a complicated (but manifestly positive) propagator and used exact

97: renormalisation group equations to estimate the ribbon graphs. The

98: technically most challenging part was a brute-force analysis

99: \cite{Grosse:2003aj} of all possible contractions of ribbon graphs.

100: The scale analysis led to the existence of an additional marginal

101: coupling in the $\phi^4$-model, which corresponds to a harmonic

102: oscillator potential for the free field. Later on, we interpreted this

103: term as required by Langmann-Szabo duality \cite{Langmann:2002cc}.  A

104: summary of these ideas can be found in \cite{Grosse:2005da}.

105: 

106: The renormalisation proof was considerably simplified by switching to

107: multi-scale analysis as the renormalisation scheme. The first version

108: still relied on the matrix base \cite{Rivasseau:2005bh}. Once the

109: bounds for the sliced propagator being proven (which is tedious), one

110: obtains in an efficient way the power-counting theorem in terms 

111: of the topology of the graph. Subsequently, the renormalisation proof

112: was also achieved by multi-scale analysis in position space (which is

113: equivalent to momentum space by Langmann-Szabo duality)

114: \cite{Gurau:2005gd}, showing the equivalence of various

115: renormalisation schemes. Recently, the position space amplitude of an

116: arbitrary orientable graph was expressed as an integral over Symanzik

117: type hyperbolic polynomials \cite{Gurau:2006yc}. With all inner

118: integrations carried out, this is the most condensed way of writing

119: Feynman graph amplitudes. See also \cite{Rivasseau:2007qx} for the

120: more complicated case of ``critical'' models.

121: 

122: Additionally, we noticed that the $\beta$-function of the

123: renormalisable noncommutative $\phi^4$-model tends to zero at large

124: energy scales. This is opposite to the commutative case and supports

125: the hope that a non-perturbative construction of the model is within

126: reach \cite{Rivasseau:2007fr,Magnen:2007uy}. The one-loop

127: $\beta$-function was first computed in \cite{Grosse:2004by} (its

128: peculiar feature was noticed in \cite{Grosse:2005da}). Roughly

129: speaking, there is a one-loop wavefunction renormalisation in the

130: model (absent in the commutative case), which for large energy scales

131: exactly compensates the renormalisation of the four-point function.

132: Then, in \cite{Disertori:2006uy} it was shown that at the self-duality

133: point $\Omega=1$ (where $\Omega$ is the frequency of the harmonic

134: oscillator potential in natural units), the $\beta$-function vanishes

135: up to three-loop order. Eventually, in \cite{Disertori:2006nq} the

136: vanishing of the $\beta$-function (at $\Omega=1$) was proven to all

137: orders, which means that the Landau ghost is absent in noncommutative

138: $\phi^4_4$-theory: Wave function renormalisation exactly compensates

139: the renormalisation of the four-point function, so that the flow

140: between the bare and the renormalised coupling is bounded.  The main

141: tool in this proof is a clever combination of the Ward identity

142: relative to unitary transformations with the Schwinger-Dyson

143: equations. Strictly speaking, the proof requires $\Omega=1$, but using

144: the bounds established in \cite{Rivasseau:2005bh}, it is plausible

145: that the renormalisation flow of the coupling is bounded for $0<\Omega<1$,

146: too. 

147: 

148: A good review of these exciting developments is

149: \cite{Rivasseau:2007ab}. The relation to previous attempts to

150: renormalise noncommutative field theories is discussed in

151: \cite{Wulkenhaar:2006si}.

152: 

153: The  importance  of  the   self-duality  case  was  first  noticed  in

154: \cite{Langmann:2003cg,Langmann:2003if} where an exact non-perturbative

155: solution of a complex scalar field theory on Moyal space with critical

156: magnetic background field was  constructed. The UV-fixed point of this

157: model is trivial.  In \cite{Grosse:2005ig,Grosse:2006tc} a non-trivial

158: exactly solvable (and just  renormalisable) field theory was obtained,

159: the noncommutative $\phi^3_6$-model at the  self-duality point. Here,

160: self-duality  relates   this  model  to   the  Kontsevich-model.   For

161: $\phi^3_4$, see \cite{Grosse:2006qv}.

162: 

163: There is also considerable progress with other than scalar field

164: models on Moyal space. In

165: \cite{VignesTourneret:2006nb,VignesTourneret:2006xa} renormalisation

166: to all orders of the duality-covariant orientable Gross-Neveu model

167: was shown. To put it into context with the work we present here, it is

168: important to stress that the Dirac operator in

169: \cite{VignesTourneret:2006nb,VignesTourneret:2006xa} is \emph{not} the

170: square root of the harmonic oscillator Hamiltonian appearing in the

171: $\phi^4$-model of \cite{Grosse:2004yu} and following treatments. It is

172: precisely in this paper where we construct such a square root and

173: analyse its properties. The Dirac operator of the Gross-Neveu model is

174: of the type studied (for scalar fields) in

175: \cite{Langmann:2003cg,Langmann:2003if}, just describing the influence

176: of a constant magnetic background field. Its spectrum is very different

177: from the harmonic oscillator (there is e.g.\ infinite degeneracy).

178: This fact can also be seen from a different structure of the

179: propagator in position space \cite{Gurau:2005qm}, which made the

180: renormalisation of the Gross-Neveu model technically more difficult.

181: In some sense, the magnetic background field is not needed for

182: renormalisation of complex scalar fields, as already argued in

183: \cite{Chepelev:2000hm} (in the massive case a new counterterm is

184: generated, though). See \cite{Lakhoua:2007ra} for the one-loop

185: $\beta$-function of this model.

186: 

187: The most interesting field theories are Yang-Mills theories, which we

188: also would like to see in renormalisable form on Moyal space. Usual

189: Yang-Mills theory on Moyal space (without modifications of the action

190: by something similar to an oscillator potential) is known to be not

191: renormalisable \cite{Matusis:2000jf}. Yang-Mills theories in

192: noncommutative geometry \cite{Connes:1994yd} are

193: naturally obtained from the spectral action principle

194: \cite{Connes:1996gi,Chamseddine:1996zu} relative to an appropriate

195: Dirac operator. In this way, a beautiful reformulation of the standard

196: model of particle physics was obtained, see \cite{Chamseddine:2006ep}

197: for its most recent version. Moyal space with undeformed Dirac

198: operator is a (non-compact) spectral triple \cite{Gayral:2003dm}. The

199: corresponding spectral action was computed in \cite{Gayral:2004ww},

200: with the result that it is the usual Yang-Mills action on Moyal space

201: (which is not renormalisable).  The magnetic background field Dirac

202: operator of the Gross-Neveu model gives the same usual Yang-Mills

203: action, too.  

204: 

205: To obtain a gauge theory with sort of oscillator potential via the

206: spectral action principle, we need a Dirac operator with similar

207: spectrum as the square root of the harmonic oscillator.

208: Unfortunately, all attempts to produce such a Dirac operator failed so

209: far, and here where we can report progress in this paper. As

210: workaround we translated the physical interpretation of the spectral

211: action (to describe a one-loop effective action of fermions in a

212: classical external gauge field) from fermions to scalar fields. In

213: \cite{Gayral:2004cs} this method was already worked out for general

214: (isospectral) Rieffel deformations \cite{Rieffel:1993??}. We finished

215: the computation almost simultaneously in position space \cite{de

216:   Goursac:2007gq} and in the matrix base \cite{Grosse:2007dm}. See

217: also \cite{Grosse:2006hh,Wallet:2007em}. As a result, there are two

218: additional terms to the Yang-Mills action, namely the integral over

219: $\tilde{X}_\mu\star \tilde{X}^\mu$ and over its square, where

220: $\tilde{X}_\mu(x)=(\Theta^{-1})_{\mu\nu} x^\nu + A_\mu(x)$ is a

221: covariant coordinate \cite{Madore:2000en}. The existence of such a

222: term was conjectured in \cite[p.\ 90]{raimar-habil}.

223: 

224: The problem with the effective action derived in \cite{de

225:   Goursac:2007gq,Grosse:2007dm} is that, expanding $\tilde{X}_\mu\star

226: \tilde{X}^\mu$ and its square, there is a linear term in the gauge

227: field $A_\mu$. The consequence is that $A_\mu=0$ \emph{is not a stable

228: solution of the classical field equation}. Any attempt to solve the

229: classical field equations resulting from \cite{de

230:   Goursac:2007gq,Grosse:2007dm} failed so far, and this is the second

231: point where we can report progress in the present paper. To circumvent

232: the vacuum problem, in \cite{Blaschke:2007vc} an oscillator potential

233: for the gauge field was achieved solely from a generalised ghost

234: sector, in a BRST-invariant way. Although a one-loop calculation is

235: likely to produce the $\tilde{X}_\mu\star \tilde{X}^\mu$ terms as in

236: \cite{de Goursac:2007gq,Grosse:2007dm}, the investigations in

237: \cite{Blaschke:2007vc} demonstrate the enormous freedom of

238: constructing the ghost sector, which in some way will be needed to

239: obtain a manageable gauge field propagator. 

240: 

241: \subsection{Strategy of the paper}

242: 

243: Our paper starts from a simple observation, so simple that it is

244: embarrassing not having it earlier exploited. The harmonic oscillator

245: Hamiltonian $H$ in one-dimensional configuration space, thus

246: two-dimensional phase space, has spectrum $\omega(n+\frac{1}{2})$ with

247: $n\in \mathbb{N}$. Thus, $H^{-1}$ is a noncommutative infinitesimal

248: \cite{Connes:1996gi} of order one---the configuration space dimension.

249: The Hamiltonian $H$ generalises the Laplacian. The central object in

250: noncommutative geometry is the Dirac operator, which is a

251: (generalised) square root of the Laplacian. Now,

252: $\mathcal{D}=H^{\frac{1}{2}}$ is a noncommutative infinitesimal of

253: order one over two, two being the phase space dimension. Spectral

254: dimension is defined through the Dirac operator so that \emph{the

255:   spectral dimension of the harmonic oscillator is the phase space

256:   dimension}.

257: 

258: For field theory we are interested in four-dimensional Moyal

259: configuration space. The isospectral deformation would be a

260: four-dimensional spectral triple \cite{Gayral:2003dm}. But for

261: renormalisation of the $\phi^4_4$-theory we must promote the 4D

262: Laplace operator $-\Delta$ to the 4D harmonic oscillator Hamiltonian

263: $H=-\Delta + \Omega^2 \|x\|^2$.  According to the previous discussion,

264: the noncommutative dimension of the 4D harmonic oscillator Hamiltonian

265: is the phase space dimension, which is EIGHT, not four. We thus

266: understand why all attempts to find a 4D Dirac operator for the 4D

267: harmonic oscillator Hamiltonian necessarily failed. On the other hand,

268: it is absolutely trivial to write down an 8D Dirac operator so that

269: its square equals (up to a constant matrix) the 4D harmonic oscillator

270: Hamiltonian. This is what we do in Section 2. Additionally, we show

271: that our 8D-Dirac operator on 4D-Moyal space almost extends to an

272: eight-dimensional spectral triple in the original sense

273: \cite{Connes:1996gi}. The orientability axiom is violated. We do not

274: check Poincar\'e duality.

275: 

276: It is worthwhile to mention that the distinction between configuration

277: space and phase space dimension was crucial for the quantum field

278: theory on projective modules over the noncommutative torus

279: investigated in \cite{Gayral:2006wu}. There, $\mathbb{R}^2$ and the

280: 2-dimensional space of holomorphic $\mathbb{C}^2$-function where

281: considered as projective modules, i.e.\ configuration space, over the

282: 4D-noncommutative torus (which extends to a four-dimensional spectral

283: triple). The resulting Hamiltonian was precisely that of the

284: $2D$-harmonic oscillator, where the oscillator potential is naturally

285: obtained from the isospectral Dirac operator of the 4D-noncommutative

286: torus.  The field theory on 2D-configuration space was shown to be

287: one-loop renormalisable like a 4D-scalar field theory, four being the

288: phase space dimension of the noncommutative torus. The dimensional

289: relations with Moyal space were discussed to some extent in

290: \cite{Gayral:2006wu}. It was noticed that the heat kernel traces split

291: into a local integral over field monomials times a \emph{partial trace

292:   only} of the propagator (see also \cite{Gayral:2004cs}).  But the

293: true dimensionality of the harmonic oscillator Moyal space was not

294: realised.

295: 

296: Having the 8D-Dirac operator with harmonic oscillator spectrum, we

297: perform the standard procedure \cite{Connes:1996gi,Chamseddine:1996zu}

298: of noncommutative geometry to get to the spectral action. To make it a

299: little more interesting, we add in Section 3 another Connes-Lott copy

300: \cite{Connes:1990qp} and compute in Section 4 the spectral action for

301: the resulting two $U(1)$-Moyal Yang-Mills fields unified with a

302: complex Higgs field to a single noncommutative gauge field. This

303: extends the computation of \cite{de Goursac:2007gq,Grosse:2007dm}

304: where the effective scalar field action was (unfortunately) not

305: considered. It turns out that only the inclusion of the Higgs field

306: provides an understanding of the $\tilde{X}_\mu\star \tilde{X}^\mu$

307: terms: We find that they appear together with the Higgs field $\phi$

308: in a potential of the form $ (\alpha \tilde{X}_\mu\star \tilde{X}^\mu

309: + \beta \phi\star \phi -1)^2$, for some positive numbers

310: $\alpha,\beta$. Thus, \emph{the origin of the non-trivial gauge field vacuum

311: is nothing but the standard Higgs mechanism}. We experience here a

312: further level of the unification of Higgs and gauge fields through

313: noncommutative geometry: Almost-commutative geometry obtained the

314: potential of the Higgs field as part of the unified Yang-Mills action.

315: Spacial noncommutativity intertwines gauge and Higgs field even

316: further so that the potential combines Higgs and gauge field on an

317: equal footing.

318: 

319: We wondered whether the combined classical field equations for gauge

320: and Higgs field (derived and investigated in Section 5) are easier to

321: solve for the vacuum than the equations without Higgs field. It turned

322: out that this is indeed the case perturbatively in the expansion

323: parameter $\frac{\Omega^2}{1+\Omega^2}$. Here, the additional Higgs

324: vacuum equation provides the necessary perturbative input for the

325: solution of the gauge field vacuum equation so that, in principle, a

326: complete perturbative solution is possible. We give the solution up to

327: order $(\frac{\Omega^2}{1+\Omega^2})^2$. From the first terms we guess

328: a general ansatz for the vacuum configuration. Inserting this ansatz

329: back into the field equations, it turns out that the Yang-Mills term

330: reduces to a modified Bessel equation for the order parameter

331: $\tilde{X}_\mu\star \tilde{X}^\mu$, so that we can solve

332: $\tilde{X}_\mu\star \tilde{X}^\mu$ in terms of the Higgs field.

333: For the time being, we cannot solve the other Higgs equation so that we

334: cannot provide here a full non-perturbative vacuum solution for both gauge

335: and Higgs field. But in the special case where the Higgs field is

336: absent, which is the type of action obtained in \cite{de

337:   Goursac:2007gq,Grosse:2007dm}, we give the explicit vacuum solution

338: for the order parameter $\tilde{X}_\mu\star \tilde{X}^\mu$ in terms of

339: modified Bessel and Struve functions.

340: 

341: 

342: 

343: 

344: 

345: 

346: 

347: 

348: 

349: 

350: 

351: \section{A spectral triple in dimension 8}

352: 

353: The renormalisable real $\phi^4$-model on the $4$-dimensional Moyal

354: plane is characterised by the appearance of the

355: harmonic oscillator Hamiltonian

356: \begin{align}

357: H_m=-\frac{\partial^2}{\partial x_\mu \partial x^\mu} +\Omega^2 \tilde{x}^\mu

358: \tilde{x}_\mu +m^2

359: \end{align}

360: in the action functional \cite{Grosse:2004yu}, where

361: $\tilde{x}_\mu:=2(\Theta^{-1})_{\mu\nu} x^\nu$. For simplicity we

362: choose

363: \begin{align}

364: \Theta=\left(\begin{array}{cccc}

365: 0 & \theta & 0 & 0 \\

366: -\theta & 0 & 0 & 0 \\

367: 0 & 0 & 0 & \theta \\

368: 0 & 0 & -\theta & 0

369: \end{array}\right)=:i\theta \sigma\;,\qquad \theta \in \mathbb{R}\;,

370: \end{align}

371: where $\sigma=\sigma_2\otimes 1_2$ consists of two copies of the

372: second Pauli matrix. We have $\Theta^{-1}=\frac{-i}{\theta} \sigma$.

373: It is then a well-known fact from quantum mechanics that the Hilbert

374: space $L^2(\mathbb{R}^4)$ has an orthonormal basis $\{

375: \psi_{\underline{n}} \}_{\underline{n} \in \mathbb{N}^4}$ of

376: eigenfunctions of $H_m$ with 

377: \[

378: H_m \psi_{\underline{n}}

379: =\frac{4\Omega}{\theta}\big( |{\underline{n}}| 

380: + \tfrac{\theta m^2}{2\Omega}\big) 

381:  \psi_{\underline{n}}\;,\qquad |{\underline{n}}|=n_1+n_2+n_3+n_4~\text{ for } 

382: ~{\underline{n}}=(n_1,n_2,n_3,n_4)\;.

383: \]

384: The inverse $H_m^{-1}$ extends to a selfadjoint compact operator on

385: $L^2(\mathbb{R}^4)$ with eigenvalues

386: \begin{align}

387: \lambda_n(m)=\Big(\frac{4\Omega}{\theta}\big( n+2 + \tfrac{\theta

388:   m^2}{2\Omega}\big)\Big)^{-1}\;,\qquad n \in \mathbb{N}\;.

389: \end{align}

390: The $n^{\mathrm{th}}$ eigenspace $E_n$ has dimension 

391: $\mathrm{dim} (E_n)= \binom{n+3}{3}$,

392: which is the number of possibilities to write $n$ as a sum of four ordered

393: natural numbers. This means that for $s>4$, the trace 

394: \begin{align}

395: \mathrm{Tr}(H_m^{-s}) = \frac{1}{6} \sum_{n=0}^\infty (n+3)(n+2)(n+1) 

396: (\lambda_n(m))^s

397: \end{align}

398: exists. The critical value $s=4$ characterises $H^{-4}$ as belonging

399: to the Dixmier trace ideal

400: $\mathcal{L}^{(1,\infty)}(L^2(\mathbb{R}^4))$ of compact operators

401: \cite{Connes:1994yd}. 

402: 

403: At first sight, $H^{-4} \in

404: \mathcal{L}^{(1,\infty)}(L^2(\mathbb{R}^4))$ seems to be related to

405: the four dimensional Moyal space under consideration.  However, recall

406: that in noncommutative geometry it is the \emph{Dirac operator} which

407: defines the dimension \cite{Connes:1996gi}. In a $d$-dimensional space

408: we require $|\mathcal{D}|^{-d} \in

409: \mathcal{L}^{(1,\infty)}(L^2(\mathbb{R}^4))$.  Identifying

410: $H=|\mathcal{D}|^2$, we notice the surprising fact that the

411: $4$-dimensional Moyal space has actually spectral dimension EIGHT.

412: 

413: In eight dimensions it is very easy to write down an appropriate Dirac

414: operator,

415: \begin{align}

416: \mathcal{D}_8= i\Gamma^\mu \partial_\mu + \Omega \Gamma^{\mu+4}

417: \tilde{x}_\mu\;. 

418: \end{align}

419: Here, the $\Gamma_k \in M_{16}(\mathbb{C})$, $k=1,\dots,8$ are the generators

420: of the 8-dimensional real Clifford algebra, satisfying

421: \begin{align}

422: \Gamma_k\Gamma_l + \Gamma_l \Gamma_k = 2\delta_{kl} 1 \;.

423: \end{align}

424: We agree

425: that latin indices run from 1 to 8 and greek indices from 1 to 4.  

426: Summation over repeated upper and lower indices is self-understood.  

427: 

428: 

429: Accordingly, we take the Hilbert space

430: $\mathcal{H}_8=L^2(\mathbb{R}^4,\mathcal{S})$ of square integrable spinors

431: over FOUR-dimensional euclidean space, where the spinor bundle

432: has typical fibre $\mathbb{C}^{16}$. For $\psi \in \mathcal{H}_8$ we obtain

433: \begin{align}

434: \mathcal{D}_8^2 \psi = \big( (-\Delta + \Omega^2 \tilde{x}_\mu 

435: \tilde{x}^\mu )1 

436: + \Sigma  \big) \psi\;,\qquad \Sigma := -i\Omega

437: (\Theta^{-1})_{\mu\nu} [\Gamma^\mu,\Gamma^{\nu+4}]\;,

438: \end{align}

439: with $\Delta=\partial_\mu \partial^\mu$.

440: Assuming a choice of the Clifford algebra where $\Sigma$ is diagonal, we

441: obtain up to the 16-fold multiplicity of each level and an unimportant shift in

442: the mass  exactly the spectrum of the harmonic oscillator Hamiltonian

443: $H$. In particular, $|\mathcal{D}_8|^{-8}$ belongs as required to the Dixmier

444: trace ideal $\mathcal{L}^{(1,\infty)}(L^2(\mathbb{R}^4,\mathcal{S}))$. 

445: 

446: As algebra $\mathcal{A}_8$ we take the unitalised Moyal

447: algebra\footnote{This choice of the algebra cannot verify the 

448:  orientability axiom in any form, because we cannot represent  

449:  the partition of unity localised at infinity (which

450:   belongs to $\mathcal{A}_8$) by derivatives of elements of the

451:   algebra (which is not possible with

452:   $\mathcal{A}_8$). This can be achieved by 

453:   an appropriate subalgebra of the multiplier algebra of

454:   $\mathbb{R}^4_\Theta$, see \cite{Gayral:2003dm}. But the

455:   orientability axiom fails anyway, so it suffices

456:   to work with $\mathcal{A}_8$.}

457: \begin{align}

458: \mathcal{A}_8 = \mathbb{R}^4_\Theta \oplus \mathbb{C}\;,

459: \end{align}

460: where $\mathbb{R}^4_\Theta$ is as a vector space given by the Schwarz

461: class functions on $\mathbb{R}^4$, equipped with the Moyal product

462: \begin{align}

463: (f\star g)(x) = \int d^4 y\,\frac{d^4 k}{(2\pi)^4}  \,f(x{+}\tfrac{1}{2}

464: \Theta \cdot k) \,g(x{+}y)\, \mathrm{e}^{i\langle k,y\rangle}\;,

465: \qquad f,g \in \mathcal{A}_8 \;.

466: \label{Moyal}

467: \end{align}

468: The Moyal product extends to constant functions using the integral

469: representation of the Dirac distribution.

470: 

471: 

472: The algebra $\mathcal{A}_8$ acts on $\mathcal{H}_8$ also by

473: componentwise Moyal product, $\star: \mathcal{A}_8 \times

474: \mathcal{H}_8 \to \mathcal{H}_8$ (we refer to \cite{Gayral:2003dm} for

475: the necessary extension of the Moyal product). Clearly, the smooth

476: spinors form a finitely generated projective module over

477: $\mathcal{A}_8$.

478: 

479: We compute the commutator of that action with the Dirac operator,

480: taking for smooth spinors the identity $2x^\mu \psi=x\star \psi + \psi

481: \star x$ into account, as well as the relation $[x^\nu ,f]_\star

482: =i\Theta^{\nu\rho} \partial_\rho f$:

483: \begin{align}

484: &\mathcal{D}_8 (f \star  \psi)- f\star (\mathcal{D}_8 \psi)

485: \nonumber

486: \\

487: &=  i\Gamma^\mu ((\partial_\mu f) \star  \psi 

488: + f \star \partial_\mu \psi)

489: + \tfrac{1}{2} \Omega \Gamma^{\mu+4}

490: (\tilde{x}_\mu \star (f\star \psi) + 

491: (f\star \psi) \star\tilde{x}_\mu )

492: \nonumber

493: \\

494: &-  i\Gamma^\mu f \star \partial_\mu \psi

495: - \tfrac{1}{2} \Omega \Gamma^{\mu+4}

496: (f\star (\tilde{x}_\mu \star \psi) +

497: f\star (\psi \star\tilde{x}_\mu ))

498: \nonumber

499: \\

500: &= \big( i(\Gamma^\mu + \Omega \Gamma^{\mu+4})

501: (\partial_\mu f) \big) \star  \psi \;.

502: \end{align}

503: Thus, just the four-dimensional differential of $f$ appears, no

504: $x$-multiplication! This differential is represented on $\mathcal

505: {H}_8$ by $\pi(dx^\mu)= \Gamma^\mu + \Omega \Gamma^{\mu+4}$, and it is

506: bounded. It commutes with Moyal multiplication from the right, so that

507: the order-one condition is achieved in the usual way.  However, the

508: algebra generated by $[\mathcal{D}_8,\mathcal{A}_8]$ and

509: $\mathcal{A}_8$ does not contain the chirality matrix $\Gamma_9$ so

510: that the orientability axiom does not hold. The ingredients of the

511: spectral triple which just rely on the Clifford algebra (dimension

512: table) are automatically satisfied. We do not check Poincar\'e

513: duality. In conclusion, up to the orientability axiom (and possibly

514: Poincar\'e duality), $(\mathcal{A}_8,\mathcal{H}_8,\mathcal{D}_8)$

515: forms a spectral triple of dimension 8.

516: 

517: 

518: \section{$U(1)$-Higgs model}

519: 

520: In the Connes-Lott spirit \cite{Connes:1990qp} we take the tensor

521: product of the 8-dimensional spectral triple

522: $(\mathcal{A}_8,\mathcal{H}_8,\mathcal{D}_8,\Gamma_9)$ with the finite

523: Higgs spectral triple $(\mathbb{C}\oplus\mathbb{C}, \mathbb{C}^2,

524: M\sigma_1)$. The Dirac operator $\mathcal{D}=\mathcal{D}_8 \otimes 1 +

525: \Gamma_9 \otimes M\sigma_1$ of the product triple becomes

526: \begin{align}

527: \mathcal{D} =\left(\begin{array}{cc} 

528: \mathcal{D}_8 & M \Gamma_9 \\ 

529: M \Gamma_9  & \mathcal{D}_8

530: \end{array}\right)\;.

531: \end{align}

532: In this representation, the algebra is $\mathcal{A}_8 \oplus \mathcal{A}_8

533: \ni (f,g)$, which acts on $\mathcal{H}=\mathcal{H}_8 \oplus \mathcal{H}_8$ by

534: diagonal Moyal multiplication. The commutator of $\mathcal{D}$ with $(f,g)$ is

535: \begin{align}

536: [\mathcal{D},(f,g)] =\left(\begin{array}{cc} 

537: i(\Gamma^\mu + \Omega \Gamma^{\mu+4})

538: L_\star(\partial_\mu f)  & M \Gamma_9 L_\star(g-f)\\ 

539: M \Gamma_9  L_\star (f-g) & i(\Gamma^\mu + \Omega \Gamma^{\mu+4})

540: L_\star(\partial_\mu g)

541: \end{array}\right)\;,

542: \end{align}

543: where $L_\star(f)\psi=f\star \psi$ is left Moyal multiplication.

544: This shows that selfadjoint fluctuated Dirac operators $\mathcal{D}_A=

545: \mathcal{D} + \sum_i a_i [\mathcal{D},b_i]$ are of the

546: form 

547: \begin{align}

548: \mathcal{D}_A  =\left(\begin{array}{cc} 

549: \mathcal{D}_8 + (\Gamma^\mu + \Omega \Gamma^{\mu+4})L_\star(A_\mu)

550:   & \Gamma_9 L_\star({\phi}) \\ \Gamma_9 L_\star(\bar{\phi})  & 

551: \mathcal{D}_8 + (\Gamma^\mu + \Omega \Gamma^{\mu+4}) L_\star(B_\mu)

552: \end{array}\right)\;,

553: \end{align}

554: for real fields $A_\mu,B_\mu \in \mathcal{A}_8$ and a complex field 

555: $\phi \in \mathcal{A}_8$. The square of $\mathcal{D}_A$ is

556: \begin{align}

557: \mathcal{D}_A^2 &= \left(\begin{array}{cc} 

558: (H_0^2+ L_\star(\phi\star \bar{\phi})) 1 +\Sigma  +F_A   & 

559: i(\Gamma^\mu + \Omega \Gamma^{\mu+4})\Gamma_9 L_\star(D_\mu \phi) 

560: \\[1ex]

561: i (\Gamma^\mu + \Omega \Gamma^{\mu+4})\Gamma_9

562: L_\star(\overline{D_\mu \phi}) & (H_0^2+ L_\star(\bar{\phi}\star \phi))1 

563: +\Sigma  +F_B

564: \end{array}\right)  \;,

565: \label{DA2}

566: \end{align}

567: where 

568: \begin{align} 

569: D_\mu \phi &:= \partial_\mu \phi -iA\star \phi+i\phi \star B\;,

570: \\[1ex]

571: F_A &:= \{\mathcal{D}_8, (\Gamma^\mu + \Omega \Gamma^{\mu+4}) L_\star(A_\mu)\} 

572: + (\Gamma^\mu + \Omega \Gamma^{\mu+4})

573: (\Gamma^\nu + \Omega \Gamma^{\nu+4})

574: L_\star(A_\mu \star A_\nu)

575: \nonumber

576: \\

577: &=  \big\{ L_\star(A^\mu) ,  i \partial_\mu + \Omega^2 

578: M_\bullet(\tilde{x}_\mu) \big\}

579: + (1+\Omega^2) L_\star(A_\mu \star A^\mu) \nonumber

580: \\

581: & 

582: + i \big(\tfrac{1}{4}

583:   [\Gamma^\mu,\Gamma^\nu] +\tfrac{1}{4} \Omega^2

584:   [\Gamma^{\mu+4},\Gamma^{\nu+4}] + \Omega \Gamma^\mu \Gamma^{\nu+4} \big)

585:   L_\star(F^A_{\mu\nu})\;,

586: \end{align}

587: and similarly for $F_B$. In this expression, 

588: $F^A_{\mu\nu} = \partial_\mu A_\nu - \partial_\nu A_\mu -i(A_\mu \star

589: A_\nu - A_\nu \star A_\mu)$ is the field strength and

590: $(M_\bullet(\tilde{x}_\mu) \psi)(x) = \tilde{x}_\mu \psi(x)$ is ordinary

591: local multiplication. 

592: 

593: 

594: \section{The spectral action}

595: 

596: \subsection{General remarks}

597: 

598: According to the spectral action principle

599: \cite{Connes:1996gi,Chamseddine:1996zu}, the bosonic action depends

600: only on the spectrum of the Dirac operator. Thus, by functional

601: calculus, the most general form of the bosonic action is

602: \begin{align}

603: S(\mathcal{D}_A)=\mathrm{Tr}\big(\chi(\mathcal{D}_A^2)\big)\;,

604: \end{align}

605: for some function $\chi:\mathbb{R}_+ \to \mathbb{R}_+$ for which the

606: Hilbert space trace exists. By Laplace transformation one has

607: \begin{align}

608: S(\mathcal{D}_A)=\int_0^\infty dt\; 

609: \mathrm{Tr}(e^{-t \mathcal{D}_A^2}) \hat{\chi}(t)\;,  

610: \end{align}

611: where $\hat{\chi}$ is the (inverse) Laplace transform of $\chi$,

612: $\chi(s)=\int_0^\infty dt\;e^{-st} \hat{\chi}(t)$. Assuming the heat

613: kernel has an asymptotic expansion

614: \begin{align}

615: e^{-t \mathcal{D}_A^2} = \sum_{n=0}^\infty a_n(\mathcal{D}_A^2) 

616: t^{-\delta +n}\;,\qquad \delta \in \mathbb{N}\;,

617: \end{align}

618: we obtain

619: \begin{align}

620: S(\mathcal{D}_A)= \sum_{n=0}^\infty \mathrm{Tr}(a_n(\mathcal{D}_A^2)) 

621: \int_0^\infty dt\; t^{-\delta +n} \hat{\chi}(t)

622: =: \sum_{n=0}^\infty \chi_n\,\mathrm{Tr}(a_n(\mathcal{D}_A^2)) \;.

623: \end{align}

624: The most singular order $\delta$ is half of the dimension according to

625: Weyl's theorem. To compute the $\chi_n$ we have to distinguish the cases

626: $-\delta +n\geq 0$ and $-\delta +n< 0$. For $n\geq \delta$ we consider

627: \begin{align}

628: \int_0^\infty \!\!dt\; t^{-\delta +n} \hat{\chi}(t)

629: &=\lim_{s\to 0} \int_0^\infty \!\! dt\; e^{-st} t^{-\delta +n} \hat{\chi}(t)

630: =\lim_{s\to 0}\; (-1)^{n-\delta} \frac{\partial^{n-\delta}}{\partial

631:   s^{n-\delta}}  \int_0^\infty dt\; e^{-st} \hat{\chi}(t)

632: \nonumber

633: \\

634: &=\lim_{s\to 0} \,(-1)^{n-\delta} \frac{\partial^{n-\delta} \chi}{\partial

635:   s^{n-\delta}}  (s)

636: =(-1)^{n-\delta}   \chi^{(n-\delta)}(0)\;.

637: \end{align}

638: For $\delta>n$ we have 

639: \begin{align}

640: \int_0^\infty ds\;s^{\delta-n-1} \chi(s) 

641: &=\int_0^\infty ds \int_0^\infty dt\;e^{-st} s^{\delta-n-1} \hat{\chi}(t) 

642: =\Gamma(\delta-n) \int_0^\infty dt\; t^{n-\delta} \hat{\chi}(t) \;.

643: \end{align}

644: In summary,

645: \begin{align}

646: \chi_n=\left\{\begin{array}{l@{\quad \text{for }}l}

647: \displaystyle \frac{1}{\Gamma(\delta-n)}\int_0^\infty

648: ds\;s^{\delta-n-1} \chi(s)

649: & \delta >n \\[2.5ex]

650: (-1)^{n-\delta}  \chi^{(n-\delta)}(0) & n\geq \delta\;.

651: \end{array}\right.

652: \end{align}

653: 

654: In a position space basis, the Hilbert space trace is given by 

655: \begin{align}

656: \mathrm{Tr}(e^{-t \mathcal{D}_A^2})=\int_{\mathbb{R}^4} dx \;

657: \mathrm{tr}\big((e^{-t \mathcal{D}_A^2})(x,x)\big)\;,

658: \end{align}

659: where $\mathrm{tr}$ denotes the matrix trace (including the Clifford

660: algebra) and $(e^{-t \mathcal{D}_A^2})(x,y)$ is the heat kernel.

661: We write

662: \begin{align}

663: \mathcal{D}_0^2:=\mathrm{H}_0\;,\qquad \mathcal{D}_A^2=:\mathcal{D}_0^2-V  

664: =\mathrm{H}_0-V\;,

665: \end{align}

666: and consider the Duhamel expansion (see \cite{Gayral:2005ih} for more

667: information) 

668: \begin{align}

669: e^{-t_0 (\mathrm{H}_0-V)} 

670: &=

671:  e^{-t_0 \mathrm{H}_0} - \int_0^{t_0}  \!\!dt_1 \;

672: \frac{d}{dt_1} \big(e^{-(t_0-t_1) (\mathrm{H}_0-V)} 

673: e^{-t_1 \mathrm{H}_0}\big) \nonumber 

674: \\

675: &=

676:  e^{-t_0 \mathrm{H}_0} + \int_0^{t_0}  \!\!dt_1 \; 

677: \big(e^{-(t_0-t_1) (\mathrm{H}_0-V)} 

678: V e^{-t_1 \mathrm{H}_0}\big) \nonumber 

679: \\

680: &= e^{-t_0 \mathrm{H}_0} + \int_0^{t_0} \!\! dt_1 \; 

681: \big(e^{-(t_0-t_1) \mathrm{H}_0} 

682: V e^{-t_1 \mathrm{H}_0}\big) \nonumber

683: \\

684: & + \int_0^{t_0} \!\! dt_1 \int_0^{t_0-t_1} \!\! dt_2 

685: \; \big(e^{-(t_0-t_1-t_2) \mathrm{H}_0} 

686: V e^{-t_2 \mathrm{H}_0}V e^{-t_1 \mathrm{H}_0}

687: \big) +\dots \nonumber 

688: \\

689: & + \int_0^{t_0} \!\! dt_1 \dots 

690: \int_0^{t_0-t_1-\dots-t_{n-1}} \!\! dt_n 

691: \; \big(e^{-(t_0-t_1-\dots-t_n) \mathrm{H}_0} 

692: (V e^{-t_n \mathrm{H}_0})\cdots (V e^{-t_1 \mathrm{H}_0})

693: \big) 

694: +\dots \nonumber

695: \\

696: &= e^{-t_0 \mathrm{H}_0} + \sum_{n=1}^\infty t_0^n \int_{\Delta^n} d^n\alpha

697: \Big(e^{-t_0(1-|\alpha|)\mathrm{H}_0}

698: \prod_{j=1}^n (V e^{-t_0\alpha_j \mathrm{H}_0})\Big) \;,

699: \label{Duhamel}

700: \end{align}

701: where the integration is performed over the standard $n$-simplex

702: $\Delta^n :=\{\alpha:=(\alpha_1,\dots,\alpha_n) \in \mathbb{R}^n\;,~

703: \alpha_i \geq 0\;,~ |\alpha|:=\alpha_1+\dots+\alpha_n \leq 1\}$.

704: 

705: 

706: \subsection{Position space kernels}

707: 

708: According to (\ref{DA2}) we have $\mathrm{H}_0=H_01_{32} + \Sigma

709: 1_2$. Its position space kernel is

710: \begin{align}

711: (e^{-t \mathrm{H}_0})(x,y)

712: &= \int d^4z \;(e^{-t H_0 1_{32}})(x,z) \;(e^{-t \Sigma 1_2})(z,y)

713: = e^{-t \Sigma 1_2} (e^{-t H_0})(x,y)\nonumber

714: \\ 

715: &= \Big(\frac{\tilde{\Omega}}{2\pi\sinh(2\tilde{\Omega} t)}

716: \Big)^2

717: e^{-t\Sigma 1_2 -\frac{\tilde{\Omega}}{4}\big(

718: \coth(\tilde{\Omega}t)|x-y|^2 

719: +\tanh(\tilde{\Omega}t) |x+y|^2 \big)}\;,

720: \end{align}

721: where the main part is given by the four-dimensional Mehler kernel

722: (see e.g.\ \cite{Simon:1979??}), with

723: $\tilde{\Omega}:=\frac{2\Omega}{\theta}$ and $|x|^2:=x_\mu x^\mu$. It

724: will be convenient to distinguish the following vertices in

725: (\ref{DA2}):

726: \begin{subequations}

727: \begin{align}

728: V_\phi &=  - L_\star(\phi \star \phi) 1_{16}\;, \\

729: V_{D\phi} &=  -i L_\star(D_\mu \phi) (\Gamma^\mu+\Omega

730: \Gamma^{\mu+4})\Gamma_9  \;,

731: \\

732: V_A &= 

733: - (1+\Omega^2) L_\star(A_\mu \star A^\mu) \;,

734: \\

735: V_{DA} &= -\big\{ L_\star(A^\mu) ,  i \partial_\mu + \Omega^2 

736: M_\bullet(\tilde{x}_\mu) \big\}\;,

737: \\

738: V_{FA} &=-

739: i L_\star(F^A_{\mu\nu})

740: \big(\tfrac{1}{4}

741:   [\Gamma^\mu,\Gamma^\nu] +\tfrac{\Omega^2}{4}

742:   [\Gamma^{\mu+4},\Gamma^{\nu+4}] 

743: + \tfrac{\Omega}{2} \Gamma^\mu \Gamma^{\nu+4} 

744: - \tfrac{\Omega}{2} \Gamma^\nu \Gamma^{\mu+4} 

745: \big)

746:   \;,

747: \end{align}

748: \end{subequations}

749: and similarly for $V_B$, $V_{DB}$ and $V_{FB}$. 

750: 

751: We compute the necessary position space kernels: 

752: \begin{align}

753: (L_\star (f) g)(x) 

754: &= \int d^4 y\,\Big( \int\frac{d^4 k}{(2\pi)^4}  \,f(x{+}\tfrac{1}{2}

755: \Theta \cdot k) \,\mathrm{e}^{i\langle k,y-x\rangle}\Big) g(y)

756: \nonumber

757: \\

758: &=   \int d^4 y\,\Big( \frac{1}{\pi^4\theta^4} \int d^4 z  \,f(z)

759: \,\mathrm{e}^{2i(\langle x,\Theta^{-1} y\rangle

760: +\langle y,\Theta^{-1} z\rangle

761: +\langle z,\Theta^{-1} x\rangle)}\Big) g(y)\;,

762: \end{align}

763: from which we get 

764: \begin{align}

765: (L_\star (f))(x,y)=\frac{1}{\pi^4\theta^4} \int d^4 z  \,f(z)

766: \,\mathrm{e}^{i\langle x-y,\Theta^{-1} (x+y)\rangle

767: +2i\langle z,\Theta^{-1} (x-y)\rangle}\;.

768: \end{align}

769: Next, we compute 

770: \begin{align}

771: &\big(\big\{ L_\star(A^\mu) , i \partial_\mu + \Omega^2 

772: M_\bullet(\tilde{x}_\mu) \big\} g\big)(x) 

773: \nonumber

774: \\

775: &=   \int d^4 y\,\Big( \frac{1}{\pi^4\theta^4} \int d^4 z  \,A^\mu(z)

776: \,\mathrm{e}^{2i(\langle x,\Theta^{-1} y\rangle

777: +\langle y,\Theta^{-1} z\rangle

778: +\langle z,\Theta^{-1} x\rangle)}\Big) 

779: \Big(i \frac{\partial g}{\partial y^\mu }(y) 

780: +  \Omega^2  \tilde{y}_\mu g(y)\Big)

781: \nonumber

782: \\

783: &+ \Big(i\frac{\partial}{\partial x^\mu} +\Omega^2 \tilde{x}_\mu\Big)

784: \Big( \int d^4 y\,\Big( \frac{1}{\pi^4\theta^4} \int d^4 z  \,A^\mu(z)

785: \,\mathrm{e}^{2i(\langle x,\Theta^{-1} y\rangle

786: +\langle y,\Theta^{-1} z\rangle

787: +\langle z,\Theta^{-1} x\rangle)}\Big) g(y)\Big)

788: \nonumber

789: \\

790: &=   \int \!\! d^4 y\,\Big( \frac{1}{\pi^4\theta^4} \int \!\! d^4z  \,

791: (2\tilde{z}^\mu -(1{-}\Omega^2)(\tilde{x}^\mu{+}\tilde{y}^\mu))A_\mu(z) 

792: \,\mathrm{e}^{2i(\langle x,\Theta^{-1} y\rangle

793: +\langle y,\Theta^{-1} z\rangle

794: +\langle z,\Theta^{-1} x\rangle)}\Big) g(y)\;.

795: \end{align}

796: Therefore, the position space kernel of a Moyal-derivative vertex is 

797: \begin{align}

798: &\big\{ L_\star(A^\mu) , i \partial_\mu + \Omega^2 

799: M_\bullet(\tilde{x}_\mu) \big\}(x,y)

800: \nonumber

801: \\

802: &=  \frac{1}{\pi^4\theta^4} \int \!\! d^4z  \,

803: (2\tilde{z}^\mu -(1{-}\Omega^2)(\tilde{x}^\mu{+}\tilde{y}^\mu))A_\mu(z) 

804: \,\mathrm{e}^{2i(\langle x,\Theta^{-1} y\rangle

805: +\langle y,\Theta^{-1} z\rangle

806: +\langle z,\Theta^{-1} x\rangle)}\;.

807: \end{align}

808: 

809: 

810: 

811: \subsection{Computation of the traces}

812: 

813: The first term in the expansion (\ref{Duhamel}), which corresponds to

814: vacuum graphs, has the heat kernel expansion

815: \begin{align}

816: \mathrm{Tr}(e^{-t \mathrm{H}_0}) &=\mathrm{tr} \int d^4x \;

817: (e^{-t \mathrm{H}_0})(x,x)

818: =  \Big(\frac{\tilde{\Omega}}{2\pi\sinh(2\tilde{\Omega} t)} \Big)^2

819: 2\,\mathrm{tr} \int d^4x \;

820: e^{-t\Sigma  -\tilde{\Omega}

821: \tanh(\tilde{\Omega}t) |x|^2} \nonumber

822: \\

823: &=  \frac{1}{8 \sinh^4(\tilde{\Omega} t)} 

824: \mathrm{tr} \big(e^{-t\Sigma}\big)\;.

825: \end{align}

826: We need the traces of the lowest powers of $\Sigma$:

827: \begin{align}

828: \mathrm{tr}(\Sigma^0)=16\;,\qquad  

829: \mathrm{tr}(\Sigma^2 )=16 \cdot \frac{16  \Omega^2}{\theta^2}\;,\qquad  

830: \mathrm{tr}(\Sigma^4)=16 \cdot \frac{160 \Omega^4}{\theta^4}\;.

831: \end{align}

832: All odd powers of $\Sigma$ are traceless. Therefore,

833: \begin{align}

834: \mathrm{Tr}(e^{-t \mathrm{H}_0}) &=

835: \frac{\theta^4}{8 \Omega^4 t^4}  + \frac{\theta^2}{\Omega^2t^2}

836: + \frac{5}{6} +\mathcal{O}(t^2)\;.

837: \end{align}

838: This reconfirms that the noncommutative space under consideration is

839: of dimension 8.

840: 

841: 

842: 

843: In the appendix we compute 

844: the first and second order $x$-$y$ integrals 

845: \begin{align}

846: & \int d^4x\, d^4y\; (e^{-tH_0})(y,x) V(x,y)\;,\\

847: & \int d^4x_1 \,d^4y_1\, d^4x_2 \,d^4y_2\; 

848: (e^{-(t-t_2)H_0})(y_2,x_1) V(x_1,y_1)

849: (e^{-t_2H_0})(y_1,x_2) V'(x_2,y_2)\;, \nonumber

850: \end{align}

851: where $V,V'$ stand for combinations of the Moyal and Moyal-derivative

852: vertices. In second order, we also perform a Taylor expansion about

853: coinciding external positions. It is remarkable that only terms of

854: order $t^{-1}$ and regular terms in $t$ appear, just as in

855: 4D-Yang-Mills theory. Only the vacuum graphs behave like a

856: 8D-model, for proper graphs only partial 4D-traces appear.

857: 

858: In the following, we only consider the trace of the $16$-dimensional

859: upper left corner containing the $A$-field and the structure

860: $\phi\star\bar{\phi}$. At the very end we add the lower right corner

861: where $A$ is replaced by $B$ and $\phi \leftrightarrow \bar{\phi}$.

862: 

863: 

864: With one $V_A$ or $V_{DA}$ vertex we see from (\ref{VLf}) and (\ref{VDA})

865: that the leading divergence after $t_1$-integration is $\sim

866: t^{-1}$. Therefore, the $\Sigma$ matrix gives no contribution up to

867: order $t^0$, so that the leading terms are 

868: \begin{align}

869: S_{(A+DA)}(t) & :=

870: \mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 \; \big(e^{-(t_0-t_1) (\mathrm{H}_0)} 

871: (V_A+V_{DA}) e^{-t_1 \mathrm{H}_0}\big)\Big)

872: \nonumber

873: \\

874: &= \frac{1}{\pi^2(1+\Omega^2)}

875: \int d^4 z \;\Big\{ 

876: -\frac{4\Omega^2}{(1+\Omega^2)} t^{-1}

877: \tilde{z}^\mu A_\mu(z)

878: +\frac{4\Omega^4}{(1+\Omega^2)^2} 

879: \tilde{z}^\mu A_\mu(z) \tilde{z}^\nu \tilde{z}_\nu

880: \nonumber

881: \\

882: & -  (1+\Omega^2) t^{-1} (A_\mu\star A^\mu)(z) 

883: +  \Omega^2 (A_\mu\star A^\mu)(z) 

884: \tilde{z}^\nu \tilde{z}_\nu \Big\} + \mathcal{O}(t)\;.

885: \end{align}

886: A single $V_{FA}$-vertex gets a non-vanishing trace of order $t^0$ 

887: together with one $\Sigma$-matrix, but the resulting integral 

888: $\int dz \;F_{\mu\nu}(z)$ vanishes. With two  $V_{FA}$-vertices and

889: the trace 

890: \begin{align}

891: &\mathrm{tr}\Big(

892: i \big(\tfrac{1}{4}

893:   [\Gamma^\mu,\Gamma^\nu] +\tfrac{1}{4} \Omega^2

894:   [\Gamma^{\mu+4},\Gamma^{\nu+4}] 

895: + \tfrac{1}{2} \Omega \Gamma^\mu \Gamma^{\nu+4}

896:  - \tfrac{1}{2}\Omega \Gamma^\nu \Gamma^{\mu+4}  \big)  

897: \nonumber

898: \\

899: & \qquad \times i \big(\tfrac{1}{4}

900:   [\Gamma^\rho,\Gamma^\sigma] +\tfrac{1}{4} \Omega^2

901:   [\Gamma^{\rho+4},\Gamma^{\sigma+4}] 

902: + \tfrac{1}{2} \Omega \Gamma^\rho \Gamma^{\sigma+4}

903:  - \tfrac{1}{2}\Omega \Gamma^\sigma \Gamma^{\rho+4}  \big)  

904:   \Big)

905: \nonumber

906: \\

907: & = 4(1-\Omega^2)^2 (\delta_{\mu\rho}\delta_{\nu\sigma}-

908: \delta_{\mu\sigma}\delta_{\nu\rho}) 

909: \end{align}

910: we find with (\ref{V2fg})

911: \begin{align}

912: S_{(FA)^2}(t) &:= \mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 

913: \int_0^{t-t_1}  \!\!dt_2 

914: \; e^{-(t_0-t_1-t_2) (\mathrm{H}_0)} 

915: V_{FA} e^{-t_2 \mathrm{H}_0}

916: V_{FA} e^{-t_1 \mathrm{H}_0}\Big)

917: \nonumber

918: \\

919: &= \frac{(1-\Omega^2)^2}{4\pi^2 (1+\Omega^2)^2} t^0

920: \int d^4z \; F^A_{\mu\nu}(z) F_A^{\mu\nu}(z)+\mathcal{O}(t)\;.

921: \end{align}

922: 

923: 

924: For two $V_{DA}$-vertices we obtain from (\ref{V2A0}) after some

925: integrations by parts

926: \begin{align}

927: S_{(DA)^2}(t) &:=\mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 

928: \int_0^{t-t_1}  \!\!dt_2 

929: \; e^{-(t_0-t_1-t_2) (\mathrm{H}_0)} 

930: V_{DA} e^{-t_2 \mathrm{H}_0}

931: V_{DA} e^{-t_1 \mathrm{H}_0}\Big)

932: \nonumber

933: \\

934: &= 16\int_0^{t}  \!\!dt_1 

935: \int_0^{t-t_1}  \!\!dt_2 

936: \; \frac{1}{(4\pi t)^2 (1+\Omega^2)^4} 

937: \int d^4 z 

938: \nonumber

939: \\

940: &

941: \times 

942: \Big( \frac{2(1-\Omega^2)^2(1+\Omega^2)}{t} A_\mu(z)A^\mu(z)

943: - 2\Omega^2 (1-\Omega^2)^2 

944: A_\mu(z)A^\mu(z) |\tilde{z}|^2

945: \nonumber

946: \\

947: &\qquad 

948: + A^\mu(z)(\partial^\nu\partial_\nu  A_\mu)(z)

949: \Big(2(1-\Omega^2)^4 \frac{t_2(t-t_2)}{t^2} 

950: +2 \Omega^2(1-\Omega^2)^2 \Big) 

951: \nonumber

952: \\

953: & +16\Omega^4 \tilde{z}^\mu  A_\mu(z)\,

954: \tilde{z}^\nu A_\nu(z)

955: + (1-\Omega^2)^4 \frac{t^2-4t_2t+4t_2^2}{t^2}

956: (\partial_\nu A_\mu)(z)\,(\partial^\mu A^\nu)(z)

957: \Big)

958: \nonumber

959: \\

960: &= \frac{1}{\pi^2 (1+\Omega^2)^2} 

961: \int d^4 z

962: \Big( \frac{(1-\Omega^2)^2}{1+\Omega^2} 

963: t^{-1} A_\mu \star A^\mu

964: \nonumber

965: \\*

966: &\qquad 

967: - \frac{(1-\Omega^2)^4}{6(1+\Omega^2)^2} \big(

968: (\partial^\nu A^\mu) \star (\partial_\nu  A_\mu)

969: - (\partial^\nu A^\mu) \star (\partial_\mu A_\nu)\big)

970: \nonumber

971: \\*

972: & \qquad - \frac{\Omega^2 (1-\Omega^2)^2}{(1+\Omega^2)^2} 

973: A_\mu \star A^\mu |\tilde{z}|^2

974: +\frac{8\Omega^4}{(1+\Omega^2)^2} 

975: \Big( (\tilde{z}\cdot A) \star (\tilde{z}\cdot A) \Big)(z)\;.

976: \end{align}

977: We have used 

978: \begin{align}

979: A^\mu(z) \tilde{z}_\nu \tilde{z}^\nu =  

980: A^\mu(z) \star (\tilde{z}_\nu \tilde{z}^\nu) 

981: + i (\partial_\nu A^\mu)(z)\tilde{z}^\nu  

982: + (\partial_\nu\partial^\nu A^\mu)(z) 

983: \end{align}

984: as well as $\int d^4z \;

985: A_\mu(z) (\partial_\nu A^\mu)(z)\,\tilde{z}^\nu=0$.

986: 

987: The $A$-linear and

988: $A$-bilinear part of the spectral action are given by  

989: the sum $S_{(A+DA)}+S_{(FA)^2}+S_{(DA)^2}$. As the spectral action is

990: manifestly gauge invariant, we simply complete the $A$-trilinear and 

991: $A$-quadrilinear terms in a gauge-invariant way. Introducing covariant

992: coordinates 

993: \begin{align}

994: \tilde{X}_{A}^\mu(z) :=\tfrac{\tilde{z}^\mu}{2}+A^\mu(z)\;,

995: \end{align}

996: with $\tilde{X}_{0}^\mu(z)= \tfrac{\tilde{z}^\mu}{2}$, 

997: we obtain the pure $A$-part of the spectral action to

998: \begin{align}

999: S_A(t) &= \frac{1}{\pi^2(1+\Omega^2)^2} 

1000: \int d^4 z  \;\Big\{ - 

1001: \frac{4\Omega^2}{1+\Omega^2}t^{-1}

1002: \big( \tilde{X}^\mu_A  \star \tilde{X}_{A\mu}

1003: - \tilde{X}^\mu_0  \star \tilde{X}_{0\mu}\big)

1004: \nonumber

1005: \\

1006: & + \frac{t^0}{2} \Big(\frac{4\Omega^2}{1+\Omega^2}\Big)^2 

1007: \big(\tilde{X}^\mu_A \star \tilde{X}_{A\mu}\star

1008: \tilde{X}^\nu_A \star \tilde{X}_{A\nu}

1009: - \tilde{X}^\mu_0 \star \tilde{X}_{0\mu}\star

1010: \tilde{X}^\nu_0 \star \tilde{X}_{0\nu}\big)

1011: \nonumber

1012: \\

1013: & + \Big(\frac{(1-\Omega^2)^2}{4}

1014: - \frac{(1-\Omega^2)^2}{6 (1+\Omega^2)^2}\Big) 

1015: t^0 \,F^A_{\mu\nu}\star  F_A^{\mu\nu}\Big\}(z)

1016: +\mathcal{O}(t)\;.

1017: \end{align}

1018: 

1019: 

1020: The scalar field potential becomes

1021: \begin{align}

1022: S_{(\phi+\phi^2)}(t) 

1023: &=\mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 \; \big(e^{-(t_0-t_1) (\mathrm{H}_0)} 

1024: V_\phi e^{-t_1 \mathrm{H}_0}\big)\Big)

1025: \nonumber

1026: \\

1027: &+ \mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 

1028: \int_0^{t-t_1}  \!\!dt_2 

1029: \; e^{-(t_0-t_1-t_2) (\mathrm{H}_0)} 

1030: V_{\phi} e^{-t_2 \mathrm{H}_0}

1031: V_{\phi} e^{-t_1 \mathrm{H}_0}\Big)

1032: \nonumber

1033: \\

1034: &=  \frac{1}{\pi^2(1+\Omega^2)^2}

1035: \int d^4z\;\Big( - t^{-1} 

1036: \phi\star \bar{\phi} 

1037: +\frac{\Omega^2|\tilde{z}|^2}{1+\Omega^2}

1038: \phi\star \bar{\phi} 

1039: +  \frac{1}{2} \phi\star \bar{\phi} \star 

1040: \phi\star \bar{\phi}\Big)(z)\;.

1041: \end{align}

1042: The usual kinetic term of the scalar field comes from two

1043: $D\phi$-vertices:

1044: \begin{align}

1045: S_{(D\phi)^2}(t) &= \mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 

1046: \int_0^{t-t_1}  \!\!dt_2 

1047: \; e^{-(t_0-t_1-t_2) (\mathrm{H}_0)} 

1048: V_{D\phi} e^{-t_2 \mathrm{H}_0}

1049: V_{\overline{D\phi}} e^{-t_1 \mathrm{H}_0}\Big)

1050: \nonumber

1051: \\

1052: &=\frac{1}{2\pi^2(1+\Omega^2)^2} t^0 \int d^4z\; \big(D_\mu\phi\star 

1053: \overline{D_\mu \phi}\big)(z) + \mathcal{O}(t)\;.

1054: \end{align}

1055: It remains the combination of $V_\phi$ with $V_A$ and $V_{DA}$, namely

1056: $V_\phi V_A$, $V_{A} V_\phi$ as well as $V_{DA} V_\phi$, $V_\phi

1057: V_{DA}$ and $V_\phi V_{DA} V_{DA}$, $V_{DA} V_\phi V_{DA}$, $V_{DA}

1058: V_{DA} V_\phi$. Up to first order in $A$ we get

1059: \begin{align}

1060: S_{(DA\phi+\phi DA)}(t)&= \mathrm{Tr}\Big( \int_0^{t}  \!\!dt_1 

1061: \int_0^{t-t_1}  \!\!\!\!dt_2 

1062: \; 

1063: \Big(e^{-(t_0-t_1-t_2) (\mathrm{H}_0)} 

1064: \Big(V_{\phi} e^{-t_2 \mathrm{H}_0}

1065: V_{A} e^{-t_1 \mathrm{H}_0}

1066: +

1067: V_{A} e^{-t_2 \mathrm{H}_0}

1068: V_{\phi} e^{-t_1 \mathrm{H}_0}\Big) \Big)

1069: \nonumber

1070: \\*

1071: &= \frac{4\Omega^2}{\pi^2(1+\Omega^2)^3} 

1072: \int d^4z\;\big(\phi\star \bar{\phi}\star (\tilde{z}^\mu

1073: A_\mu)\big)(z)+\mathcal{O}(t)\;.

1074: \end{align}

1075: Completing the the $AA\phi\bar{\phi}$-term by gauge invariance, the

1076: scalar field part of the spectral action becomes 

1077: \begin{align}

1078: S_\phi(t) = \frac{1}{\pi^2(1+\Omega^2)^2}&

1079: \int d^4z\;  \Big( -t^{-1}\phi\star \bar{\phi} 

1080: +\frac{1}{2} D_\mu\phi\star \overline{D_\mu \phi}

1081: \nonumber

1082: \\

1083: & + \frac{1}{2} \phi\star\bar{\phi}\star \Big(

1084: \phi\star\bar{\phi} +2\frac{4\Omega^2}{1+\Omega^2} 

1085: \tilde{X}^\mu_A \star \tilde{X}_{A\mu}\Big)\Big)(z)\;.

1086: \end{align}

1087: 

1088: To obtain the spectral action, we convert the Laplace-transform 

1089: variable $t^{n-4}$ into $\chi_n$ and add the lower $B$-corner. The result

1090: (including the vacuum contribution is 

1091: \begin{align}

1092: S&= \frac{\theta^4 \chi_0}{8 \Omega^4}  + \frac{\theta^2 \chi_2 }{\Omega^2}

1093: + \frac{5 \chi_4}{6} 

1094: \nonumber

1095: \\

1096: & + \frac{\chi_4}{2\pi^2(1+\Omega^2)^2}

1097: \int d^4z\;  \Big\{

1098: \big(\tfrac{(1-\Omega^2)^2}{2}

1099: - \tfrac{(1-\Omega^2)^4}{3 (1+\Omega^2)^2}\big) \,

1100: (F^A_{\mu\nu}\star  F_A^{\mu\nu}+F^B_{\mu\nu}\star  F_B^{\mu\nu})

1101: \nonumber

1102: \\

1103: & +\Big(\phi\star \bar{\phi} 

1104: + \tfrac{4\Omega^2}{1+\Omega^2} \tilde{X}^\mu_A \star 

1105: \tilde{X}_{A\mu} -\frac{\chi_3}{\chi_4}\Big)^2 

1106: +\Big(\bar{\phi} \star \phi

1107: + \tfrac{4\Omega^2}{1+\Omega^2} \tilde{X}^\mu_B \star 

1108: \tilde{X}_{B\mu} -\frac{\chi_3}{\chi_4}\Big)^2 

1109: \nonumber

1110: \\

1111: & 

1112: -2  \Big(\tfrac{4\Omega^2}{1+\Omega^2}

1113: \tilde{X}^\mu_0 \star \tilde{X}_{0\mu} -\frac{\chi_3}{\chi_4}\Big)^2

1114: +2 D_\mu\phi\star \overline{D_\mu \phi}

1115: \Big\}(z)  +\mathcal{O}(\chi_{5})\;.

1116: \label{SpecAct}

1117: \end{align}

1118: The most important conclusion is that the squared covariant

1119: derivatives combine with the Higgs field to a non-trivial

1120: potential. This was not noticed in \cite{de

1121:   Goursac:2007gq,Grosse:2007dm}. 

1122: 

1123: 

1124: \section{Classical field equations}

1125: 

1126: The appearance of $\tilde{X}^\mu_A \star 

1127: \tilde{X}_{A\mu}$ and its square leads to a non-vanishing vacuum

1128: expectation value of the gauge field. Further progress with 

1129: noncommutative gauge theories was obstructed by the impossibility to

1130: construct this vacuum explicitly. We now turn to the classical field

1131: equations in the hope that the inclusion of the Higgs field helps to

1132: identify the vacuum. 

1133: 

1134: To simplify the notation, we

1135: let $\frac{1}{4g^2} := \tfrac{(1-\Omega^2)^2}{2} -

1136: \tfrac{(1-\Omega^2)^4}{3 (1+\Omega^2)^2}$ and

1137: $\frac{\chi_3}{\chi_4}=\eta^2$.  Then, by variational principle we get from

1138: the spectral action (\ref{SpecAct}) the following set of classical

1139: field equations:

1140: \begin{align}

1141: D^\mu D_\mu \phi &= 

1142: 2 \phi \star ( \bar{\phi}\star \phi-\eta^2) 

1143: + \tfrac{4\Omega^2}{1+\Omega^2} \big( \tilde{X}^\mu_A

1144: \star \tilde{X}_{A\mu} \star \phi +  \phi \star \tilde{X}^\mu_B \star 

1145: \tilde{X}_{B\mu}\big)\;,

1146: \\

1147: g^{-2} D^\nu_A F^A_{\nu\mu} &= 2i \big(

1148: D_\mu \phi \star \bar{\phi} -\phi \star \overline{D_\mu \phi}\big)

1149: + \frac{8\Omega^2}{1+\Omega^2} 

1150: \Big(

1151: \tilde{X}_{A\mu} \star \big( \phi\star \bar{\phi}

1152: + \tfrac{4\Omega^2}{1+\Omega^2} \tilde{X}^\nu_A \star 

1153: \tilde{X}_{A\nu} -\eta^2\big)

1154: \nonumber

1155: \\*

1156: &\qquad\qquad 

1157: +\big( \phi\star \bar{\phi}

1158: + \tfrac{4\Omega^2}{1+\Omega^2} \tilde{X}^\nu_A \star 

1159: \tilde{X}_{A\nu} -\eta^2\big)\star \tilde{X}_{A\mu}\Big)\;,

1160: \\

1161: g^{-2} D^\nu_B F^B_{\nu\mu} &= 2i \big(

1162: \overline{D_\mu \phi} \star \phi -\bar{\phi} \star D_\mu \phi\big)

1163: + \frac{8\Omega^2}{1+\Omega^2} 

1164: \Big(

1165: \tilde{X}_{B\mu} \star \big( \bar{\phi}\star \phi

1166: + \tfrac{4\Omega^2}{1+\Omega^2} \tilde{X}^\nu_B \star 

1167: \tilde{X}_{B\nu} -\eta^2\big)

1168: \nonumber

1169: \\*

1170: &\qquad\qquad 

1171: +\big( \bar{\phi}\star \phi

1172: + \tfrac{4\Omega^2}{1+\Omega^2} \tilde{X}^\nu_B \star 

1173: \tilde{X}_{B\nu} -\eta^2\big)\star \tilde{X}_{B\mu}\Big)\;.

1174: \end{align}

1175: The solution can be obtained as a formal power series in 

1176: $\frac{\Omega^2}{1+\Omega^2}$,

1177: \begin{align}

1178: \phi&=\eta + \sum_{j=1}^\infty 

1179: \Big(\frac{\Omega^2}{1+\Omega^2}\Big)^j \phi^{(j)} \;, & 

1180: A_\mu&=  \sum_{j=1}^\infty 

1181: \Big(\frac{\Omega^2}{1+\Omega^2}\Big)^j A_\mu^{(j)} \;, & 

1182: B_\mu&=  \sum_{j=1}^\infty 

1183: \Big(\frac{\Omega^2}{1+\Omega^2}\Big)^j B_\mu^{(j)} \;.

1184: \end{align}

1185: 

1186: \subsection{First order}

1187: 

1188: We obtain

1189: \begin{align}

1190: (\partial^\mu\partial_\mu \phi^{(1)})(x) -i \eta 

1191: (\partial^\mu A^{(1)}_\mu-\partial^\mu B^{(1)}_\mu)(x)

1192: &= 2 \eta^2

1193: (\phi^{(1)}(x)+\overline{\phi^{(1)}(x)})

1194: + \frac{8\eta}{\theta^2} |x|^2

1195: \end{align}

1196: with particular solution

1197: \begin{align}

1198: \phi^{(1)}(x) = -\frac{2|x|^2}{\eta \theta^2}

1199: -\frac{4}{\eta^3\theta^2}\;,\qquad 

1200: \partial^\mu A^{(1)}_\mu=\partial^\mu B^{(1)}_\mu\;.

1201: \end{align}

1202: This solution is real, which implies

1203: \begin{subequations}

1204: \begin{align}

1205: \partial^\nu(\partial_\nu A^{(1)}_\mu-\partial_\mu A^{(1)}_\nu) &= 

1206: 4g^2\eta^2 (A^{(1)}_\mu-B^{(1)}_\mu) \;, 

1207: \\

1208: \partial^\nu(\partial_\nu B^{(1)}_\mu-\partial_\mu B^{(1)}_\nu) &= 

1209: 4g^2\eta^2 (B^{(1)}_\mu-A^{(1)}_\mu) 

1210:   \end{align}

1211: \end{subequations}

1212: with particular solution $A^{(1)}_\mu=B^{(1)}_\mu=0$.

1213: 

1214: 

1215: \subsection{Second Order}

1216: 

1217: In second order we get 

1218: \begin{align}

1219: (\partial^\mu\partial_\mu \phi^{(2)})(x) 

1220: -i \eta 

1221: (\partial^\mu A^{(2)}_\mu-\partial^\mu B^{(2)}_\mu)(x)

1222: &= 6 \eta (\phi^{(1)}\star

1223: \phi^{(1)})(x) + 2 \eta^2 

1224: (\phi^{(2)}(x)+\overline{\phi^{(2)}(x)})

1225: \nonumber

1226: \\

1227: &+ \frac{4}{\theta^2} (|x|^2 \star \phi^{(1)}

1228: + \phi^{(1)} \star |x|^2 ) (x)\;.

1229: \end{align}

1230: Using $|x|^2\star |x|^2=(|x|^2)^2-2\theta^2$, we obtain   

1231: $\partial^\mu\partial_\mu \phi^{(2)}

1232: = \frac{8(|x|^2)^2}{\eta\theta^4} 

1233: -\frac{16}{\eta\theta^2} 

1234: +\frac{64|x|^2}{\eta^3\theta^4} 

1235: +\frac{96}{\eta^5\theta^4} 

1236: +4\eta^2 \phi^{(2)}$ 

1237: with particular solution

1238: \begin{align}

1239: \phi^{(2)}(x)= - \frac{2(|x|^2)^2}{\eta^3\theta^4} 

1240: -\frac{28 |x|^2}{\eta^5 \theta^4} -\frac{80}{\eta^7\theta^4}

1241: +\frac{4}{\eta^3\theta^2} \;.

1242: \end{align}

1243: For the gauge fields we obtain

1244: \begin{subequations}

1245: \begin{align}

1246: \partial^\nu(\partial_\nu A^{(2)}_\mu-\partial_\mu A^{(2)}_\nu) &= 

1247: 2g^2 i (\partial_\mu \phi^{(1)} \star \phi^{(1)} -

1248: \phi^{(1)} \star \partial_\mu \phi^{(1)}) 

1249: \nonumber

1250: \\

1251: &+ 4g^2\eta^2 (A^{(2)}_\mu-B^{(2)}_\mu) 

1252: +8 \tilde{x}_\mu\Big(2\eta \phi^{(1)} +4 \frac{|x|^2}{\theta}\Big)

1253: \;, 

1254: \\

1255: \partial^\nu(\partial_\nu B^{(2)}_\mu-\partial_\mu B^{(2)}_\nu) &= 

1256: 2g^2 i (\partial_\mu \phi^{(1)} \star \phi^{(1)} -

1257: \phi^{(1)} \star \partial_\mu \phi^{(1)}) 

1258: \nonumber

1259: \\

1260: &+ 4g^2\eta^2 (B^{(1)}_\mu-A^{(1)}_\mu) 

1261: +8 \tilde{x}_\mu\Big(2\eta \phi^{(1)} +4 \frac{|x|^2}{\theta}\Big)\;.

1262:   \end{align}

1263: \end{subequations}

1264: A particular solution is

1265: \begin{align}

1266: A^{(2)}_\mu = B^{(2)}_\mu  = -\frac{8 g^2|x|^2}{\eta^2\theta^2} 

1267: \tilde{x}_\mu\;.

1268: \end{align}

1269: 

1270: 

1271: \subsection{General ansatz}

1272: 

1273: Our findings suggest to look for a real solution

1274: \begin{align}

1275: \phi=\bar{\phi}=f(|x^2|) \;,\qquad A_\mu=B_\mu=\frac{1}{2}\tilde{x}_\mu

1276: (\xi(|x|^2)-1)  

1277: \label{ansatz}

1278: \end{align} 

1279: in terms of real functions $f,\xi$ of the squared radius. This also

1280: implies $D_\mu \phi=\overline{D_\mu\phi}$. 

1281: We thus expect a unique polynomial solution 

1282: \begin{align}

1283:   \phi &= \bar{\phi}= \eta + \sum_{j=1}^\infty

1284:   \Big(\frac{\Omega^2}{1+\Omega^2}\Big)^j f^{(j)}(|x|^2) \;, & 

1285: A_\mu&=B_\mu=  \frac{\Omega^2\tilde{x}_\mu}{1+\Omega^2} \sum_{j=1}^\infty 

1286: \Big(\frac{\Omega^2}{1+\Omega^2}\Big)^j \xi^{(j)}(|x|^2) 

1287: \end{align}

1288: of the classical field equations, where $f^{(j)}$ and $\xi^{(j)}$ are

1289: polynomials of order $j$. 

1290: 

1291: As such a solution is not very explicit, we present a different

1292: treatment of the vacuum sector. Instead of inserting (\ref{ansatz})

1293: into the field equations, we first rewrite the action in terms of

1294: $f,\xi$. For this we need some properties of $\star$-products of

1295: radial functions. The easiest way is to derive them first for

1296: polynomials using the asymptotic expansion of the $\star$-product

1297: \begin{align}

1298: (a\star b)(x) & = a(|x|^2)b(|x|^2) \\

1299: &+ \sum_{n=1}^\infty \frac{1}{n!} 

1300: \Big(\frac{i}{2} \Theta^{\mu_1\nu_1}

1301: \frac{\partial}{\partial x^{\mu_1}}

1302: \frac{\partial}{\partial y^{\nu_1}}\Big)\dots

1303: \Big(\frac{i}{2} \Theta^{\mu_n\nu_n}

1304: \frac{\partial}{\partial x^{\mu_n}}

1305: \frac{\partial}{\partial y^{\nu_n}}\Big)a(|x|^2) b(|y|^2)\Big|_{x=y}\;,

1306: \nonumber

1307: \end{align}

1308: which is exact for polynomials. 

1309: 

1310: First, the $\star$-product of real radial polynomials is again a

1311: radial polynomial: Applying to $b(|y|^2)$ the $y$-derivatives together

1312: with $\Theta$ up to some order yields a polynomial of

1313: $\tilde{y}^{\mu_i}$ and $\delta^{\mu_i\mu_j}$. Applying to $a(|x|^2)$

1314: the $x$-derivatives gives a polynomial in $x_{\mu_i}$ and

1315: $\delta_{\mu_i\mu_j}$. The contraction at $x=y$ gives a polynomial in

1316: $|x|^2$ due to $\tilde{x}_\mu x^\mu=0$. Moreover, because of

1317: $\tilde{x}_\mu x^\mu=0$ there is always an even number of

1318: $\Theta$-factors. This means that $(a\star b)(|x|^2)$ is real and

1319: $a\star b=b\star a$. Another way to see this is to use the matrix base

1320: $(f_{mn})$, see e.g.\ the appendix of

1321: \cite{Grosse:2003nw,raimar-habil}. Then, a real function of the radius

1322: is necessarily diagonal.

1323: 

1324: In the same way we obtain

1325: \begin{subequations}

1326: \begin{align}

1327: (\tilde{X}_\mu \star \tilde{X}^\mu)(|x|^2) = \frac{|x|^2}{\theta^2}

1328: (\xi \star \xi)(|x|^2)\;,\\

1329: D_\mu \phi= -i[\tilde{X}_\mu,\phi]_\star =

1330: -\frac{i}{2}[\tilde{x}_\mu \xi, f]_\star = 2 x_\mu (\xi \star

1331: f')(|x|)^2\;,

1332: \\

1333: F_{\mu\nu}= -i[\tilde{X}_\mu,\tilde{X}_\nu]_\star +\Theta_{\mu\nu} =

1334: -\frac{i}{4}[\tilde{x}_\mu \xi, \tilde{x}_\nu \xi]_\star +\Theta_{\mu\nu}\;,

1335: \label{Fxi}

1336: \end{align}

1337: \end{subequations}

1338: with $f'(|x|^2)=\frac{d f}{d(|x|^2)}(|x|^2)$ being the derivative with

1339: respect to the squared radius. This is also a derivative of the

1340: $\star$-product of radial functions, 

1341: \begin{align}

1342: 2|x|^2 (a\star b)'(|x|^2) &= x^\mu \partial_\mu (a\star b) 

1343: = x^\mu ((\partial_\mu a)\star b) + 

1344: x^\mu (a \star (\partial_\mu  b)) 

1345: \nonumber

1346: \\

1347: &=  2 x^\mu ( (x_\mu a') \star b + 

1348: 2 x^\mu ( a \star (x_\mu b')) = 2 |x|^2 (a'\star b + a \star b')\;.

1349: \end{align}

1350: 

1351: In the action we can ignore the $\Theta_{\mu\nu}$-term in (\ref{Fxi}),

1352: because $\int d^4x\, F_{\mu\nu}(x)=0$ and a constant does not

1353: contribute to the field equations. We have

1354: \begin{align}

1355: -\frac{i}{4}[\tilde{x}_\mu \xi, \tilde{x}_\nu \xi]_\star (x)

1356: = \Theta^{-1}_{\nu\mu} (\xi \star \xi)(|x|^2) 

1357: +(\tilde{x}_\nu x_\mu-\tilde{x}_\mu x_\nu)  (\xi\star \xi')(|x|^2) \;.

1358: \end{align}

1359: Up to total derivatives, we have 

1360: $(|x|^2)^n a(|x|^2) b(|x|^2) = 

1361: (|x|^2)^n (a \star b)(|x|^2) +\partial_\mu q^\mu$ and 

1362: then, as part of $F_{\mu\nu} F^{\mu\nu}$, 

1363: \[

1364: \frac{4}{\theta^2} \xi \star \xi\star \xi \star \xi 

1365: + \frac{8 |x|^2}{\theta^2}\xi' \star \xi\star \xi \star \xi 

1366: = \frac{4}{\theta^2} \xi \star \xi\star \xi \star \xi 

1367: + \frac{x^\mu}{\theta^2} \partial_\mu (\xi \star \xi\star \xi \star \xi)

1368: = \partial_\mu \Big(\frac{x^\mu}{\theta^2} 

1369: (\xi \star \xi\star \xi \star \xi)\Big)\;.

1370: \]

1371: 

1372: 

1373: Then, up to field-independent terms and total derivatives, we have 

1374: \begin{align}

1375: S_{vac}=   

1376: \frac{\chi_4}{2\pi^2(1+\Omega^2)^2}

1377: \int &d^4z\;  \Big\{

1378: 2 \big( f\star f + \tfrac{4\Omega^2 |x|^2}{\theta^2 (1+\Omega^2)} \xi\star \xi

1379: -\eta^2\big)^2

1380: \nonumber

1381: \\

1382: & +\frac{4}{\theta^2 g^2} \,(|x|^2)^2 (\xi\star \xi')^2 

1383: +8 |x|^2 (\xi\star \xi \star f' \star f') \Big\}(z) \;.

1384: \label{SpecActfg}

1385: \end{align}

1386: Taking $2|x|^2 a'=x^\mu \partial_\mu a$ into account, the field

1387: equations are 

1388: \begin{align}

1389: 4 \xi\star\xi\star f' + 2|x|^2 (\xi \star \xi\star f')' 

1390: &=f \star \big( f\star f + \tfrac{4 |x|^2\Omega^2}{\theta^2(1+\Omega^2)} 

1391: \xi\star \xi -\eta^2\big)\;, 

1392: \label{feq-1}

1393: \\

1394: (|x|^2)^2 \xi \star (\xi \star \xi')'+3 |x|^2  

1395: \xi \star \xi \star \xi'

1396: &= 2\theta^2 g^2 |x|^2 \xi \star f' \star f'

1397: \nonumber

1398: \\

1399: &+\tfrac{4 \Omega^2 |x|^2 g^2}{1+\Omega^2} \xi \star 

1400: \big( f\star f + \tfrac{4|x|^2\Omega^2}{\theta^2(1+\Omega^2)} \xi\star \xi

1401: -\eta^2\big)\;.

1402: \label{feq-2}

1403: \end{align}

1404: One not so interesting solution is $\xi=0$, $f=\eta$. For $\xi \neq 0$

1405: we can replace (\ref{feq-2}) by 

1406: \begin{align}

1407: |x|^2 (\xi \star \xi)''+3  (\xi \star \xi)'

1408: - \tfrac{32 |x|^2 g^2 \Omega^4}{\theta^2 (1+\Omega^2)^2} \xi\star \xi

1409: &= 4\theta^2 g^2 f' \star f'

1410: +\tfrac{8\Omega^2 g^2}{1+\Omega^2} ( f\star f -\eta^2)\;.

1411: \label{feq-2a}

1412: \end{align}

1413: % We can also obtain a quadratic relation in terms of $f\star

1414: % f= v$ and $\xi\star \xi=\zeta$ if 

1415: % we multiply the first equation by $f$, the second by

1416: % $\frac{\xi}{g^2\theta^2}$, and take the difference: 

1417: % \begin{align}

1418: % &2 \zeta\star v' + \frac{1}{2}|x|^2 \zeta'  \star v'

1419: % + |x|^2 \zeta \star v'' 

1420: % -\frac{1}{2\theta^2 g^2} (|x|^2)^2 \zeta \star

1421: % \zeta''-\frac{3}{2\theta^2 g^2} |x|^2  

1422: % \zeta \star \zeta' 

1423: % \nonumber

1424: % \\

1425: % &=(v - \tfrac{4 \Omega^2 x^2}{\theta^2(1+\Omega^2)} \zeta)

1426: % \big( v + \tfrac{4 x^2\Omega^2}{\theta^2(1+\Omega^2)} 

1427: % \zeta -\eta^2\big)\;.

1428: % \end{align}

1429: The homogeneous equation $|x|^2 (\xi \star \xi)''+3  (\xi \star \xi)'

1430: - \tfrac{32 |x|^2 g^2 \Omega^4}{\theta^2 (1+\Omega^2)^2} \xi\star

1431: \xi=0$ has a solution $(\xi \star \xi)_0$ in terms of modified Bessel functions

1432: (see e.g.\ \cite{eom})

1433: \begin{align}

1434: (\xi \star \xi)_0(|x|^2) =\frac{c_1}{|x|^2}

1435: I_1\Big(\gamma |x|^2\Big) 

1436: + \frac{c_2}{|x|^2}

1437: K_1\Big(\gamma |x|^2\Big)

1438: \;,\qquad \gamma:= \frac{4\sqrt{2}g\Omega^2}{\theta(1+\Omega^2)}\;.

1439: \label{Bessel-hom}

1440: \end{align}

1441: To obtain a particular solution of the inhomogeneous equation, the

1442: usual strategy is to build a Green's function out of the two

1443: fundamental solutions. With $I_1'(\rho) K_1(\rho) - K_1'(\rho)

1444: I_1(\rho)= \frac{1}{\rho}$ and the differential equation of the

1445: modified Bessel functions, the Green's function is given by

1446: \begin{align}

1447: G(|x|^2,r^2)= \left\{ \begin{array}{c@{\quad \text{for }}c}  

1448: -\frac{I_1(\gamma |x|^2)}{|x|^2} \,r^2 K_1(\gamma r^2) & 

1449: |x|^2 \leq r^2 \;,\\[1ex]

1450: -\frac{K_1(\gamma |x|^2)}{|x|^2} \,r^2 I_1(\gamma r^2) & 

1451: |x|^2 \geq r^2 \;.

1452:   \end{array}\right.

1453: \end{align}

1454: Therefore, a particular solution of (\ref{feq-2a}) is given by

1455: \begin{align}

1456: (\xi \star \xi)(|x|^2) = &-4 g^2 \frac{K_1(\gamma |x|^2)}{|x|^2} 

1457: \int_0^{|x|^2} dr^2\;r^2 I_1(\gamma r^2)

1458: \Big(\theta^2 f' \star f'

1459: +\tfrac{2\Omega^2}{1+\Omega^2} ( f\star f -\eta^2)

1460: \Big)(r^2)

1461: \nonumber

1462: \\*

1463: & -4 g^2 \frac{I_1(\gamma |x|^2)}{|x|^2} 

1464: \int_{|x|^2}^\infty dr^2\;r^2 K_1(\gamma r^2)

1465: \Big(\theta^2 f' \star f'

1466: +\tfrac{2\Omega^2}{1+\Omega^2} ( f\star f -\eta^2)

1467: \Big)(r^2)\;.

1468: \label{Green}

1469: \end{align}

1470: This solution is regular. As the general solution (\ref{Bessel-hom}) 

1471: of (\ref{feq-2a}) is singular for $|x|^2 \to0$ or $|x|^2\to\infty$, we 

1472: have to regard (\ref{Green}) as the most general solution of the

1473: second field equation. 

1474: 

1475: In principle, one would insert (\ref{Green}) into the first field

1476: equation (\ref{feq-1}), which becomes a complicated

1477: integro-differential equation. There is a special (but interesting)

1478: case where we obtain a complete solution of the field equations: for

1479: absent scalar fields $f=0$ as considered in 

1480: \cite{de Goursac:2007gq,Grosse:2007dm}.  Using

1481: $I_0'(\rho)=I_1(\rho)$ and $K_0'(\rho)=-K_1(\rho)$ and manipulations

1482: of the power series of the modified Bessel function, the solution for

1483: the order parameter $\tilde{X}^\mu\star \tilde{X}_\mu=

1484: \frac{|x|^2}{\theta^2} \xi\star \xi$ is then

1485: \begin{align}

1486: (\tilde{X}^\mu\star \tilde{X}_\mu)(x)\Big|_{f=0} =

1487: &\frac{\eta^2(1{+}\Omega^2)}{4\Omega^2}    

1488: \Big(1 - 

1489: K_1(\gamma |x|^2)

1490: \int_0^{\gamma |x|^2} \!\!\!\!\! d \rho\; I_0(\rho)

1491: + I_1(\gamma |x|^2)

1492: \int_{\gamma |x|^2}^\infty  \!\!\!\! d \rho \; K_0(\rho)\Big)

1493: \nonumber

1494: \\

1495: &= \frac{\eta^2(1{+}\Omega^2)}{4\Omega^2}    

1496: \Big(1+\frac{\pi}{2}\big(I_1(\gamma |x|^2)-\mathbf{L}_{-1}(\gamma

1497: |x|^2)\big)\Big) \;.

1498: \label{Green-0}

1499: \end{align}

1500:  See Figure~\ref{fig}. Here,

1501: \begin{figure}

1502: \includegraphics[bb=0 580 300 720]{bessel-8d.ps}  

1503: \caption{The function $1+\frac{\pi}{2}\big(I_1(x)-\mathbf{L}_{-1}(x)\big)$. }

1504: \label{fig}

1505: \end{figure}

1506: \begin{align}

1507: I_\nu(z)&=

1508: \sum_{k=0}^\infty

1509: \frac{(\tfrac{z}{2})^{2k+\nu}}{\Gamma(k+1)\Gamma(k+\nu+1 )} \;, & 

1510: \mathbf{L}_{-\nu}(z)&=

1511: \sum_{k=0}^\infty

1512: \frac{(\tfrac{z}{2})^{2k-\nu+1}}{\Gamma(k+\frac{3}{2})

1513: \Gamma(k-\nu+\frac{3}{2})} 

1514: \end{align}

1515: are the modified Bessel and Struve functions, respectively.

1516: Both have a very similar asymptotic

1517: behaviour. For large arguments, one has \cite{eom}

1518: \begin{align}

1519:   I_\nu(z) - \mathbf{L}_{-\nu}(z) \approx 

1520: \frac{1}{(\frac{z}{2})^{1+\nu}

1521:   \sqrt{\pi}\Gamma(\frac{1}{2}-\nu)}\;,\qquad z\to \infty\;.

1522: \end{align}

1523: Actually, this solution (\ref{Green-0}) can immediately be obtained

1524: from the fact that the modified Struve function $\mathbf{L}_n$ solves the

1525: inhomogeneous modified Bessel equation \cite{eom}. The solution of the

1526: homogeneous equation, $I_n$, has to be added in order to cancel the

1527: singularity at $\infty$.

1528: 

1529: The order parameter $(\tilde{X}^\mu\star \tilde{X}_\mu)$ increases

1530: monotonously with

1531: \begin{align}

1532: (\tilde{X}^\mu\star \tilde{X}_\mu)(x) \stackrel{|x|\to 0}{\longrightarrow}

1533: \frac{\pi}{2\sqrt{2}} g  \eta^2 \frac{|x|^2}{\theta}

1534: \end{align}

1535: for small $|x|$ and approaches for large $|x|$ its asymptotic value 

1536: where the potential term vanishes,

1537: \begin{align}

1538: (\tilde{X}^\mu\star \tilde{X}_\mu)(x) \stackrel{|x|\to

1539:   \infty}{\longrightarrow} 

1540: \frac{\eta^2(1+\Omega^2)}{4\Omega^2}\big(1-(\gamma|x|^2)^{-2}\big)

1541: \end{align}

1542: 

1543: 

1544: 

1545: \section{Outlook}

1546: 

1547: To obtain directly the vacuum of the gauge potential $A_\mu$ or of the

1548: covariant coordinate $\tilde{X}_\mu$, one should compute the matrix

1549: base coefficients of $(\xi\star\xi)(x)=\sum_{m\in

1550:   \mathbb{N}^2}(\xi\star\xi)_m f_{mm}(x)$ and take the square root,

1551: $\tilde{X}^{vac}_\mu= \frac{\tilde{x}^\mu}{2} \sum_{m\in \mathbb{N}^2}

1552: \sqrt{(\xi\star\xi)_m} f_{mm}(x)$. This will be done elsewhere. One

1553: should then write the gauge field as

1554: $\tilde{X}_{A\mu}=\tilde{X}^{vac}_\mu+A^q_\mu$, where $A^q_\mu$ is the

1555: quantum gauge field with vanishing vacuum expectation value. Its gauge

1556: transformation is obtained from $\tilde{X}^{vac}_\mu+A^q_\mu \mapsto

1557: u\star (\tilde{X}^{vac}_\mu+A^q_\mu) \star u^{-1}$. The structure of

1558: the vacuum suggests to use a moving frame as follows. At position

1559: $x=(x_1,x_2,x_3,x_4)$ we take the orthonormal frame

1560: \begin{align}

1561: e_1 &=\tfrac{1}{\sqrt{x_1^2+x_2^2}}  (x_1,x_2,0,0)\;, &

1562: e_2 &=\tfrac{1}{\sqrt{x_1^2+x_2^2}} (-x_2,x_1,0,0)\;, \nonumber \\

1563: e_3 &=\tfrac{1}{\sqrt{x_3^2+x_4^2}}  (0,0,x_3,x_4)\;, &

1564: e_4 &=\tfrac{1}{\sqrt{x_3^2+x_4^2}} (0,0,-x_4,x_3)\;.

1565: \end{align}

1566: Then, the vacuum has (with $\tilde{x}$) only components in the

1567: $e_2,e_4$-plane, which should considerably simplify the analysis.  In

1568: this frame, we also reconfirm our ansatz for the vacuum: The

1569: $e_1,e_3$-compenents of the gauge field do not appear linearly in

1570: $\tilde{X}_\mu \star \tilde{X}^\mu$ so that their vacuum expectation

1571: value vanishes. Then, rotational invariance implies that the vacuum is

1572: of the form $A^{vac}_\mu=(e_2+e_4)_\mu

1573: \zeta(|x|^2)=\frac{\tilde{x}_\mu}{|\tilde{x}|} \zeta(|x|^2) $, for

1574: some function $\zeta$ of the radius.

1575: 

1576: Another interesting case is a pure Higgs model where $\xi=1$.

1577: 

1578: 

1579: 

1580: 

1581: \renewcommand{\thesection}{\Alph{section}}

1582: \setcounter{section}{0}

1583: 

1584: \section{Appendix: Moyal integrals}

1585: 

1586: \allowdisplaybreaks[4]

1587: 

1588: \subsection{One Moyal vertex}

1589: 

1590: We compute a generic trace term with a change of variables 

1591: $u=x-y$, $v=x+y$ with Jacobian $\frac{1}{16}$ (see \cite{Gurau:2005gd}):

1592: \begin{align}

1593: V_1(f)&:=\int d^4x \,d^4y\; (e^{-tH_0})(y,x) (L_\star(f))(x,y)

1594: \nonumber

1595: \\*

1596: &= \frac{\tilde{\Omega}^2}{4\pi^2\sinh^2(2\tilde{\Omega} t)} 

1597: \frac{1}{(2\pi\theta)^4} \int d^4 z  \,f(z)

1598: \int d^4u\, d^4v\;

1599: e^{-\frac{\tilde{\Omega}}{4}(\frac{ |u|^2}{\tanh(\tilde{\Omega}t)}

1600: +\frac{|v|^2}{\coth(\tilde{\Omega}t)})

1601: +i\langle u,\Theta^{-1} (v-2z)\rangle}

1602: \nonumber

1603: \\

1604: &= \frac{1}{\cosh^2(\tilde{\Omega} t)} 

1605: \frac{1}{(2\pi\theta)^4} \int d^4 z  \,f(z)

1606: \int d^4v\;

1607: e^{-\frac{\tilde{\Omega}\tanh(\tilde{\Omega}t)}{4}

1608: (|v|^2+ \frac{4}{\tilde{\Omega}^2\theta^2}|v+2z|^2)}

1609: \nonumber

1610: \\

1611: &= \frac{1}{\cosh^2(\tilde{\Omega} t)} 

1612: \frac{1}{(2\pi\theta)^4} \int d^4 z  \,f(z)

1613: \int d^4v\;

1614: e^{-\frac{\tanh(\tilde{\Omega}t)}{2\theta \Omega}

1615: ((1+\Omega^2) |v+\frac{2}{1+\Omega^2}z|^2 +

1616: \frac{4\Omega^2}{1+\Omega^2} |z|^2)}

1617: \nonumber

1618: \\

1619: &= \frac{\tilde{\Omega}^2}{16\pi^2 (1+\Omega^2)^2 

1620: \sinh^2(\tilde{\Omega} t)} 

1621: \int d^4 z  \,f(z)

1622: e^{-\frac{\tilde{\Omega} \tanh(\tilde{\Omega}t)}{1+\Omega^2} |z|^2}\;.

1623: \label{VLf}

1624: \end{align}

1625: 

1626: 

1627: 

1628: \subsection{One Moyal+derivative vertex}

1629: 

1630: After a change of variables 

1631: $u=x-y$, $v=x+y$ with Jacobian $\frac{1}{16}$, we have

1632: \begin{align}

1633: V_1(A) &:=\int d^4x d^4y\; (e^{-tH_0})(y,x) \big\{ L_\star(A^\mu) , 

1634: i \partial_\mu + \Omega^2 M_\bullet(\tilde{x}_\mu) \big\}(x,y)

1635: \nonumber

1636: \\

1637: &= \frac{\tilde{\Omega}^2}{4\pi^2\sinh^2(2\tilde{\Omega} t)} 

1638: \frac{1}{(2\pi\theta)^4} 

1639: \int d^4u \,d^4v \,d^4 z  \,A_\mu(z)(2\tilde{z}^\mu 

1640: - (1-\Omega^2)\tilde{v}^\mu)

1641: \nonumber

1642: \\

1643: & \times 

1644: e^{-\frac{\tilde{\Omega}}{4}(\frac{ |u|^2}{\tanh(\tilde{\Omega}t)}

1645: +\frac{|v|^2}{\coth(\tilde{\Omega}t)})

1646: +i\langle u,\Theta^{-1} (v-2z)\rangle}

1647: \nonumber

1648: \\

1649: &= \frac{\tilde{\Omega}^2}{4\pi^2\sinh^2(2\tilde{\Omega} t)} 

1650: \frac{1}{(2\pi\theta)^4} 

1651: \int d^4u \Big(\int d^4v \, d^4 z  \,A_\mu(z)\Big(2\tilde{z}^\mu 

1652: + 2i(1-\Omega^2)\frac{\partial}{\partial w_\mu}\Big)

1653: \nonumber

1654: \\

1655: & \times 

1656: e^{-\frac{\tilde{\Omega}}{4}(\frac{ |u|^2}{\tanh(\tilde{\Omega}t)}

1657: +\frac{|v|^2}{\coth(\tilde{\Omega}t)})

1658: +i\langle w,\Theta^{-1} v\rangle

1659: -2i\langle u,\Theta^{-1} z\rangle}\Big|_{w=u}\Big)

1660: \nonumber

1661: \\

1662: &= \frac{1}{\sinh^4(\tilde{\Omega} t)} 

1663: \frac{1}{(2\pi\theta)^4} 

1664: \int d^4u \Big(\int d^4 z  \,A_\mu(z)\Big(2\tilde{z}^\mu 

1665: - \frac{2i(1-\Omega^2)}{\Omega \theta \tanh(\tilde{\Omega} t)} w^\mu\Big)

1666: \nonumber

1667: \\

1668: & \times 

1669: e^{-\frac{\Omega |u|^2 }{2\theta \tanh(\tilde{\Omega}t)}

1670: -\frac{|w|^2}{2\Omega \theta \tanh(\tilde{\Omega}t)}

1671: -2i\langle u,\Theta^{-1} z\rangle}\Big|_{w=u}\Big)

1672: \nonumber

1673: \\

1674: &= 

1675: \frac{1}{(2\pi\theta)^4} 

1676: \int d^4u\,d^4 z  \,A_\mu(z)\Big(2\tilde{z}^\mu 

1677: - \frac{(1-\Omega^2)}{\Omega \theta \tanh(\tilde{\Omega} t)}

1678: \Theta^{\mu\nu}

1679: \frac{\partial}{\partial z^\nu}\Big)

1680: \frac{e^{-\frac{(1+\Omega^2) |u|^2 }{2\Omega \theta \tanh(\tilde{\Omega}t)}

1681: -2i\langle u,\Theta^{-1} z\rangle}}{\sinh^4(\tilde{\Omega} t)} 

1682: \nonumber

1683: \\

1684: &= 

1685: \frac{\Omega^2}{(\pi\theta)^2(1+\Omega^2)^2} 

1686: \int d^4 z  \,A_\mu(z)\Big(2\tilde{z}^\mu 

1687: - \frac{(1-\Omega^2)}{\Omega \theta \tanh(\tilde{\Omega} t)}

1688: \Theta^{\mu\nu}

1689: \frac{\partial}{\partial z^\nu}\Big)

1690: \frac{e^{-\frac{\tilde{\Omega} \tanh(\tilde{\Omega} t)}{(1+\Omega^2)}

1691:     |z|^2 }}{\sinh^2(2\tilde{\Omega} t)} 

1692: \nonumber

1693: \\

1694: &= 

1695: \frac{4\Omega^4}{(\pi\theta)^2(1+\Omega^2)^3\sinh^2(2\tilde{\Omega} t)} 

1696: \int d^4 z  \,\tilde{z}^\mu A_\mu(z)\, 

1697: e^{-\frac{\tilde{\Omega} \tanh(\tilde{\Omega} t)}{1+\Omega^2} |z|^2 }\;.

1698: \label{VDA}

1699: \end{align}

1700: 

1701: 

1702: This term gives the complete $A$-linear part. It vanishes for

1703: $\Omega=0$, as expected. 

1704: 

1705: 

1706: 

1707: 

1708: \subsection{Two Moyal vertices}

1709: 

1710: To simplify the notations in this case we let 

1711: $\tau_1 := \tanh(\tilde{\Omega} (t-t_2))$ and 

1712: $\tau_2 := \tanh(\tilde{\Omega} t_2)$.

1713: The change of variables $u_i=x_i-y_i$ and $v_i=x_i+y_i$ for $i=1,2$

1714: leads to

1715: \begin{align}

1716: V_2(f,g)&:=\int d^4x_1 \,d^4y_1\, d^4x_2 \,d^4y_2\; 

1717: (e^{-(t-t_2)H_0})(y_2,x_1) (L_\star(f))(x_1,y_1)

1718: \nonumber

1719: \\*

1720: &\qquad\qquad \times (e^{-t_2H_0})(y_1,x_2) (L_\star(g))(x_2,y_2)

1721: \nonumber

1722: \\

1723: &=  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1724: 16\pi^2\tau_1\tau_2}\Big)^2

1725: \frac{1}{(2\pi\theta)^8}

1726: \int d^4 u_1 \,d^4 v_1\, d^4 u_2 \,d^4v_2\,d^4z_1 \,d^4z_2\; 

1727: f(z_1)g(z_2)

1728: \nonumber

1729: \\

1730: & \times 

1731: e^{

1732: -\frac{\tilde{\Omega}}{16 \tau_1} |u_1+v_1+u_2-v_2|^2

1733: -\frac{\tilde{\Omega}\tau_1}{16} |u_1+v_1-u_2+v_2|^2

1734: -\frac{\tilde{\Omega}}{16 \tau_2} |u_1-v_1+u_2+v_2|^2

1735: -\frac{\tilde{\Omega}\tau_2}{16} |-u_1+v_1+u_2+v_2|^2}

1736: \nonumber

1737: \\

1738: & \times 

1739: e^{i\langle u_1,\Theta^{-1} v_1\rangle

1740: -2i\langle u_1,\Theta^{-1} z_1\rangle 

1741: + i\langle u_2,\Theta^{-1} v_2\rangle

1742: -2i\langle u_2,\Theta^{-1} z_2\rangle }\;.

1743: \end{align}

1744: Defining 

1745: \begin{align}

1746: C &:=  \left(\begin{array}{cccc} 

1747: 1{+}\tau_1\tau_2 & 1{-}\tau_1\tau_2 &

1748: -\frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) &

1749: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) 

1750: \\

1751: 1{-}\tau_1\tau_2 & 1{+}\tau_1\tau_2 &

1752: -\frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) &

1753: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) 

1754: \\

1755: -\frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) &

1756: -\frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) &

1757: 1{+}\tau_1\tau_2 & -(1{-}\tau_1\tau_2) 

1758: \\

1759: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) &

1760: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) &

1761: -(1{-}\tau_1\tau_2) & 1{+}\tau_1\tau_2 

1762: \end{array}\right)

1763: \nonumber

1764: \\

1765: G &:= \left(\begin{array}{cccc}   

1766: 0 & 0 & \!\!\! -\frac{4\tau_1\tau_2}{\Omega(\tau_1+\tau_2)}\!\!\! & 0 \\

1767: 0 & 0 & 0 &\!\!\! -\frac{4\tau_1\tau_2}{\Omega(\tau_1+\tau_2)} \\

1768: \frac{4\tau_1\tau_2}{\Omega(\tau_1+\tau_2)}\!\!\! & 0 & 0 & 0 \\  

1769: 0 & \!\!\!\frac{4\tau_1\tau_2}{\Omega(\tau_1+\tau_2)}\!\!\! & 0 & 0  \\  

1770: \end{array}\right)\,,\quad

1771: X:= \left(\begin{array}{c} 

1772: u_1 \\ u_2 \\ v_1 \\ v_2

1773: \end{array}\right),\quad 

1774: Z:= \left(\begin{array}{c} 

1775: z_1 \\ z_2 \\ 0 \\ 0

1776: \end{array}\right),

1777: \end{align}

1778: and $Q:= C \otimes 1_4 + G \otimes \sigma$, we obtain 

1779: \begin{align}

1780: V_2(f,g)

1781: &=  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1782: 16\pi^2\tau_1\tau_2}\Big)^2

1783: \frac{1}{(2\pi\theta)^8}

1784: \int d^{16} X d^8 Z\; 

1785: f(z_1)g(z_2)

1786: \nonumber

1787: \\

1788: & \times 

1789: e^{

1790: -\frac{\Omega (\tau_1+\tau_2)}{8\theta \tau_1\tau_2} 

1791: X^t Q X 

1792: - \frac{2}{\theta} X^t \sigma Z }\;.

1793: \nonumber

1794: \\

1795: &=  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1796: 16\pi^2\tau_1\tau_2}\Big)^2

1797: \Big(\frac{4\tau_1\tau_2}{\Omega (\tau_1+\tau_2)}\Big)^8

1798: \det( C \otimes 1_4 + G \otimes \sigma)^{-\frac{1}{2}}

1799: \nonumber

1800: \\

1801: & \times 

1802: \int d^8 Z\; 

1803: f(z_1)g(z_2)\;e^{

1804: -\frac{8\tau_1\tau_2}{\Omega\theta (\tau_1+\tau_2)}

1805: Z^t \sigma (C \otimes 1_4 + G \otimes \sigma)^{-1}  \sigma Z }\;.

1806: \end{align}

1807: In \cite{Gurau:2006yc} it was proven that

1808: \begin{align}

1809: \det (Q) &= \det (G+C)^4\;,\qquad \\

1810: Q^{-1} &= 

1811: \frac{1}{2} \big((G+C)^{-1} + ((G+C)^{-1})^t\big) \otimes 1_4

1812: +  \frac{1}{2} \big((G+C)^{-1} - ((G+C)^{-1})^t\big) \otimes \sigma\;.

1813: \end{align}

1814: We find 

1815: \begin{align}

1816: &\det (G+C) = \Big(\frac{16(1+\Omega^2)

1817:   \tau_1^2\tau_2^2}{\Omega^2 (\tau_1+\tau_2)^2}\Big)^2 

1818: \Big(1 + \frac{\Omega^2(\tau_1-\tau_2)^2}{(1+\Omega^2)^2\tau_1\tau_2}\Big)\;,

1819: \end{align}

1820: which suggests to introduce 

1821: \begin{align}

1822: T:= 1 +

1823: \frac{\Omega^2(\tau_1-\tau_2)^2}{(1+\Omega^2)^2\tau_1\tau_2}\;, 

1824: \end{align}

1825: and further

1826: \begin{align}

1827: &\frac{1}{2} \big((G+C)^{-1} + ((G+C)^{-1})^t\big)

1828: =\frac{\Omega^2(\tau_1 + \tau_2)^2}{

1829: 16 \tau_1^2 \tau_2^2 (1 + \Omega^2)T}

1830: \nonumber

1831: \\

1832: &\times \left(\!\! \begin{array}{cccc} 

1833: 1{+}\tau_1\tau_2 & -(1{-}\tau_1\tau_2) &

1834: \frac{1-\Omega^2}{1+\Omega^2} 

1835: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) &

1836: \frac{1-\Omega^2}{1+\Omega^2} 

1837: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) 

1838: \\

1839: -(1{-}\tau_1\tau_2) & 1{+}\tau_1\tau_2 &

1840: -\frac{1-\Omega^2}{1+\Omega^2} 

1841: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) &

1842: -\frac{1-\Omega^2}{1+\Omega^2} 

1843: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) 

1844: \\

1845: \frac{1-\Omega^2}{1+\Omega^2} 

1846: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) &

1847: -\frac{1-\Omega^2}{1+\Omega^2} 

1848: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) &

1849: 1{+}\tau_1\tau_2 & 1{-}\tau_1\tau_2 

1850: \\

1851: \frac{1-\Omega^2}{1+\Omega^2} 

1852: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{+}\tau_1\tau_2) &

1853: -\frac{1-\Omega^2}{1+\Omega^2} 

1854: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} (1{-}\tau_1\tau_2) &

1855: 1{-}\tau_1\tau_2 & 1{+}\tau_1\tau_2 

1856: \end{array}\!\! \right)

1857: \end{align}

1858: as well as 

1859: \begin{align}

1860: &\frac{1}{2} \big((G+C)^{-1} - ((G+C)^{-1})^t\big)

1861: =\frac{\Omega^2(\tau_1 + \tau_2)^2}{

1862: 8 \tau_1^2 \tau_2^2 (1 + \Omega^2)^2T}

1863: \nonumber

1864: \\*

1865: &\times \left(\begin{array}{cccc} 

1866: 0 & \Omega(\tau_1-\tau_2) &  \!\!\!\!\frac{2(1+\Omega^2)\tau_1\tau_2

1867:   +\Omega^2(\tau_1-\tau_2)^2}{\Omega (\tau_1+\tau_2)} \!\!\!\! & 0 \\

1868:  \!\!\!\!- \Omega(\tau_1-\tau_2) & 0 & 0  &\frac{2(1+\Omega^2)\tau_1\tau_2

1869:   +\Omega^2(\tau_1-\tau_2)^2}{\Omega (\tau_1+\tau_2)} \!\!\!\! \\

1870: - \frac{2(1+\Omega^2)\tau_1\tau_2+\Omega^2(\tau_1-\tau_2)^2}{

1871: \Omega (\tau_1+\tau_2)} \!\!\!\! & 0 & 0 & 

1872: \Omega(\tau_1-\tau_2) \\

1873: 0 &  \!\!\!\!- \frac{2(1+\Omega^2)\tau_1\tau_2+\Omega^2(\tau_1-\tau_2)^2}{

1874: \Omega (\tau_1+\tau_2)}  \!\!\!\!& -\Omega(\tau_1-\tau_2) & 0

1875: \end{array}\right).

1876: \end{align}

1877: 

1878: We thus conclude 

1879: \begin{align}

1880: V_2(f,g) &= 

1881:   \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1882: 16(1+\Omega^2)^2 \pi^2 T  \tau_1\tau_2}\Big)^2

1883: \int d^4z_1 d^4 z_2 \; f(z_1)g(z_2)\;

1884: \nonumber

1885: \\

1886: & \times 

1887: e^{-\frac{\Omega(\tau_1 + \tau_2)}{2 \theta \tau_1 \tau_2 (1 +

1888: \Omega^2)T}(|z_1-z_2|^2+\tau_1\tau_2|z_1+z_2|^2)

1889: -\frac{2\Omega^2(\tau_1^2 - \tau_2^2)}{\theta

1890: \tau_1 \tau_2 (1 + \Omega^2)^2T} z_1\sigma z_2}\;.

1891: \end{align}

1892: For $\tau_1,\tau_2 \to 0$ the integrand is regular unless

1893: $z_1=z_2$. To capture the singularity at $z_1=z_2$, we expand 

1894: $g(z_2)=g(z_1)+(z_2-z_1)\int d\xi \;(\partial_\mu g)(z_1+\xi(z_2-z_1))$

1895: and consider the leading term $g(z_1)$. After a shift $z_2\mapsto

1896: z_2+z_1$ we have

1897: \begin{align}

1898: V_2(f,g)^0 &= 

1899:   \Big(\frac{\tilde{\Omega}(1-\tau_1^2)(1-\tau_2^2)}{

1900: 4\pi(1+\Omega^2)  (\tau_1+\tau_2)(1+\tau_1\tau_2)}\Big)^2

1901: \int d^4z_1 \; f(z_1)g(z_1)\;

1902: e^{-\frac{\tilde{\Omega}(\tau_1 + \tau_2)}{(1+ \tau_1 \tau_2) (1 +

1903: \Omega^2)}|z_1|^2}\;.

1904: \label{V2fg}

1905: \end{align}

1906: It can be shown that $(z_2-z_1)\int d\xi \;(\partial_\mu

1907: g)(z_1+\xi(z_2-z_1))$ is subleading.

1908: 

1909: 

1910: \subsection{Two Moyal-derivative vertices}

1911: 

1912: To complete the $A$-bilinear part, we also need the contribution with

1913: two vertices of Moyal+derivative type. We use as far as possible the

1914: same notation as in the previous calculation. Defining the auxiliary

1915: vector $W=(0,0,w_3,w_4)^t$, this gives 

1916: \begin{align}

1917:   V_2(A,A)&:=\int d^4x_1 \,d^4y_1\, d^4x_2 \,d^4y_2\;

1918:   (e^{-(t-t_2)H_0})(y_2,x_1) \big\{ L_\star(A^\mu) , i \partial_\mu +

1919:   \Omega^2 M_\bullet(\tilde{x}_\mu) \big\}(x_1,y_1) \nonumber

1920:   \\

1921:   & \times (e^{-t_2H_0})(y_1,x_2) \big\{ L_\star(A^\mu) , i

1922:   \partial_\mu + \Omega^2 M_\bullet(\tilde{x}_\mu) \big\}(x_2,y_2)

1923:   \nonumber

1924:   \\

1925: &=

1926:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1927: 16\pi^2\tau_1\tau_2}\Big)^2

1928: \frac{1}{(2\pi\theta)^8}

1929: \int d^{16} X d^8 Z\; 

1930: A_\mu(z_1)A_\nu(z_2)

1931: \nonumber

1932: \\

1933: &\times 

1934:  \Big(2 \tilde{z}_1^\mu  -(1-\Omega^2) (\Theta^{-1})^{\mu\rho} 

1935: \frac{\partial}{\partial w_3^\rho}\Big) \Big(2

1936:   \tilde{z}_2^\nu   -(1-\Omega^2) (\Theta^{-1})^{\nu\sigma} 

1937: \frac{\partial}{\partial w_4^\sigma}\Big)

1938: \nonumber

1939: \\

1940: & \times 

1941: e^{-\frac{\Omega (\tau_1+\tau_2)}{8\theta \tau_1\tau_2} 

1942: X^t Q X 

1943: - \frac{2}{\theta} X^t \sigma Z +2 X^t W}\Big|_{W=0}

1944: \nonumber

1945: \\

1946: &=

1947:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1948: 16\pi^2\tau_1\tau_2}\Big)^2

1949: \Big(\frac{4 \tau_1\tau_2} {\Omega (\tau_1+\tau_2)}\Big)^8

1950: (\det Q)^{-1/2}

1951: \int d^8 Z\; A_\mu(z_1)A_\nu(z_2)

1952: \nonumber

1953: \\

1954: &\times 

1955: \Big(\frac{8\theta\tau_1\tau_2}{\Omega (\tau_1+\tau_2)}

1956: (1-\Omega^2)^2 (\Theta^{-1})^{\mu\rho} 

1957: (\Theta^{-1})^{\nu\sigma} \big((Q^{-1})^{43}_{\sigma\rho}+ 

1958: (Q^{-1})^{34}_{\rho\sigma}\big) 

1959: \nonumber

1960: \\

1961: &+

1962:  \Big(2 \tilde{z}_1^\mu  +\frac{8i\theta\tau_1\tau_2(1-\Omega^2)}{

1963: \Omega(\tau_1+\tau_2)} (\Theta^{-1})^{\mu\rho} 

1964: (Q^{-1} \tilde{Z})_\rho^3\Big) 

1965: \nonumber

1966: \\

1967: & \qquad \times \Big(2  \tilde{z}_2^\nu  

1968: +\frac{8i\theta \tau_1\tau_2(1-\Omega^2)}{

1969: \Omega(\tau_1+\tau_2)} (\Theta^{-1})^{\nu\sigma} 

1970: (Q^{-1} \tilde{Z})_\sigma^4\Big) 

1971: \Big)\nonumber

1972: \\

1973: & \times 

1974: e^{-\frac{8\tau_1\tau_2}{\Omega\theta (\tau_1+\tau_2)}

1975: (  Z(Q^{-1})Z + i\theta^2  W Q^{-1} \tilde{Z} -\theta^2 

1976: W Q^{-1} W )}

1977: \Big|_{W=0}

1978: \nonumber

1979: \\

1980: &=

1981:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

1982: 16(1+\Omega^2)^2\pi^2 T \tau_1\tau_2}\Big)^2

1983: \int d^4 z_1 \,d^4z_2\; A_\mu(z_1)A_\nu(z_2)

1984: \nonumber

1985: \\

1986: &\times 

1987: \Big(\frac{\Omega(1-\Omega^2)^2 

1988: (\tau_1 + \tau_2)(1-\tau_1\tau_2)}{\theta

1989: (1 + \Omega^2)T\tau_1 \tau_2 }

1990: \delta^{\mu\nu}

1991: \nonumber

1992: \\

1993: & 

1994: + \Big(  -

1995: \frac{i\tilde{\Omega}(\tau_1 - \tau_2)}{2\tau_1 \tau_2T}

1996: \frac{(1-\Omega^2)^2}{(1+\Omega^2)^2} 

1997: \big((z_1^\mu-z_2^\mu)

1998: - \tau_1\tau_2(z_1^\mu +z_2^\mu)\big) 

1999: +\frac{\Omega^2(\tau_1+\tau_2)^2}{

2000: (1+\Omega^2)\tau_1\tau_2 T} \tilde{z}_1^\mu

2001: \Big)

2002: \nonumber

2003: \\

2004: & \times \Big(

2005: -

2006: \frac{i\tilde{\Omega}(\tau_1 - \tau_2)}{2\tau_1 \tau_2 T}

2007: \frac{(1-\Omega^2)^2}{(1+\Omega^2)^2} 

2008: \big((z_1^\nu-z_2^\nu)+\tau_1\tau_2(z_1^\nu+z_2^\nu)\big)

2009: + \frac{\Omega^2(\tau_1+\tau_2)^2}{

2010: (1+\Omega^2)\tau_1\tau_2 T} \tilde{z}_2^\nu\Big)\Big)

2011: \nonumber

2012: \\

2013: & \times e^{-\frac{\Omega(\tau_1 + \tau_2)}{2 \theta \tau_1 \tau_2 (1 +

2014: \Omega^2)T}(|z_1-z_2|^2+\tau_1\tau_2|z_1+z_2|^2)

2015: -\frac{2\Omega^2(\tau_1^2 - \tau_2^2)}{\theta

2016: \tau_1 \tau_2 (1 + \Omega^2)^2T} z_1\sigma z_2}\;.

2017: \end{align}

2018: We write $(z_1-z_2) \pm\tau_1\tau_2(z_1+z_2)$ as derivative of the

2019: exponential, plus appropriate corrections, and integrate by parts:

2020: \begin{align}

2021: V_2(A,A)&=

2022:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2023: 16(1+\Omega^2)^3\pi^2 T \tau_1\tau_2}\Big)^2

2024: \int d^4 z_1 \,d^4z_2\; A_\mu(z_1)A_\nu(z_2)

2025: \nonumber

2026: \\

2027: &\times 

2028: \Big(

2029: \frac{2\tilde{\Omega}(1-\Omega^2)^2(1+\Omega^2) (1-\tau_1\tau_2)}{

2030: (\tau_1+\tau_2)}\delta_{\mu\nu}

2031: - 2i \Omega^2(1-\Omega^2)^2

2032: \frac{(\tau_1^2-\tau_2^2)}{\tau_1\tau_2 T} (\Theta^{-1})^{\mu\nu} 

2033: \nonumber

2034: \\

2035: & 

2036: + \Big(  -i(1-\Omega^2)^2

2037: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} \frac{\partial}{\partial z_{2\mu}}

2038: +4\Omega^2 \tilde{z}_1^\mu\Big)

2039: \Big(

2040: i(1-\Omega^2)^2

2041: \frac{\tau_1-\tau_2}{\tau_1+\tau_2}  \frac{\partial}{\partial z_{1\nu}}

2042: + 4\Omega^2\tilde{z}_2^\nu\Big)\Big)

2043: \nonumber

2044: \\

2045: & \times e^{-\frac{\Omega(\tau_1 + \tau_2)}{2 \theta \tau_1 \tau_2 (1 +

2046: \Omega^2)T}(|z_1-z_2|^2+\tau_1\tau_2|z_1+z_2|^2)

2047: -\frac{2\Omega^2(\tau_1^2 - \tau_2^2)}{\theta

2048: \tau_1 \tau_2 (1 + \Omega^2)^2T} z_1\sigma z_2}

2049: \nonumber

2050: \\

2051: &= \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2052: 16(1+\Omega^2)^3\pi^2 T \tau_1\tau_2}\Big)^2

2053: \int d^4 z_1 \,d^4z_2\; 

2054: \nonumber

2055: \\

2056: &

2057: \times \Big( \frac{2\tilde{\Omega}(1-\Omega^2)^2(1+\Omega^2) 

2058: (1-\tau_1\tau_2)}{\tau_1+\tau_2} A_\mu(z_1)A^\mu(z_2)

2059: +16\Omega^4 \tilde{z}_1^\mu  A_\mu(z_1)\,\tilde{z}_2^\nu A_\nu(z_2)

2060: \nonumber

2061: \\

2062: & + (1-\Omega^2)^4 \frac{(\tau_1-\tau_2)^2}{(\tau_1+\tau_2)^2} 

2063: (\partial_\nu A_\mu)(z_1)\,(\partial_\mu A_\nu)(z_2)

2064: \nonumber

2065: \\

2066: &+ 4i \Omega^2(1-\Omega^2)^2

2067: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} 

2068: \big( 

2069: A^\mu(z_1) \tilde{z}_2^\nu (\partial_\mu A_\nu)(z_2)

2070: -\tilde{z}_1^\mu (\partial_\nu A_\nu)(z_1)\,A^\nu(z_2) \big)

2071: \nonumber

2072: \\

2073: &-  2i \Omega^2(1-\Omega^2)^2

2074: \frac{\tau_1-\tau_2}{\tau_1+\tau_2}

2075: \Big(4+ \frac{(\tau_1+\tau_2)^2}{\tau_1\tau_2 T}\Big)

2076: (\Theta^{-1})^{\mu\nu} A_\mu(z_1) A_\nu(z_2)

2077: \Big)\nonumber

2078: \\

2079: & \times e^{-\frac{\tilde{\Omega}(\tau_1 + \tau_2)}{4 \tau_1 \tau_2 (1 +

2080: \Omega^2)T}(|z_1-z_2|^2+\tau_1\tau_2|z_1+z_2|^2)

2081: -\frac{2i\Omega^2(\tau_1^2 - \tau_2^2)}{

2082: \tau_1 \tau_2 (1 + \Omega^2)^2T} \langle z_1,\Theta^{-1} z_2\rangle}\;.

2083: \end{align}

2084: Again, the integrand is regular for $z_1\neq z_2$, so that we expand 

2085: \begin{align}

2086: A_\nu(z_2) &= A_\nu(z_1)

2087: +(z_2^\rho-z_1^\rho)(\partial_\rho A_\nu)(z_1)

2088: +\frac{1}{2}(z_2^\rho-z_1^\rho)(z_2^\sigma-z_1^\sigma)

2089: (\partial_\rho\partial_\sigma A_\nu)(z_1) 

2090: \nonumber

2091: \\

2092: & + \frac{1}{2}(z_2^\rho{-}z_1^\rho)(z_2^\sigma{-}z_1^\sigma)

2093: (z_2^\kappa{-}z_1^\kappa)\int_0^1 d\xi\;(1{-}\xi)^2

2094: (\partial_\rho\partial_\sigma\partial_\kappa A_\nu)(z_1{+}\xi(z_2{-}z_1)) \;,

2095: \end{align}

2096: and similarly for $(\partial_\mu A_\nu)(z_2)$. In leading $t$-order,

2097: we must develop $A_\mu(z_1)A^\mu(z_2)$ up to second order (due to the

2098: appearance of $(\tau_1+\tau_2)^{-1}$) and all other terms only up to

2099: zeroth order. These leading terms become after a shift $z_2\mapsto

2100: z_2+z_1$

2101: \begin{align}

2102: V_2(A,A)^0 &= \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2103: 16(1+\Omega^2)^3\pi^2 T \tau_1\tau_2}\Big)^2

2104: \int d^4 z_1 \,d^4z_2\; 

2105: \nonumber

2106: \\

2107: &

2108: \times \Big( \frac{2\tilde{\Omega}(1-\Omega^2)^2(1+\Omega^2) 

2109: (1-\tau_1\tau_2)}{\tau_1+\tau_2} \Big(

2110: A_\mu(z_1)A^\mu(z_1)+ A_\mu(z_1)(\partial_\rho A^\mu)(z_1)

2111: \frac{\partial}{\partial w_\rho}

2112: \nonumber

2113: \\ &\qquad + 

2114: \frac{1}{2} A_\mu(z_1)(\partial_\rho\partial_\sigma  A^\mu)(z_1)

2115: \frac{\partial^2}{\partial w_\rho \partial w_\sigma}\Big)

2116: \nonumber

2117: \\

2118: & +16\Omega^4 \tilde{z}_1^\mu  A_\mu(z_1)\,

2119: \tilde{z}_1^\nu A_\nu(z_1)

2120: + (1-\Omega^2)^4 \frac{(\tau_1-\tau_2)^2}{(\tau_1+\tau_2)^2} 

2121: (\partial_\nu A_\mu)(z_1)\,(\partial_\mu A_\nu)(z_1)

2122: \nonumber

2123: \\

2124: &+ 2(\Theta^{-1})^{\nu\rho} \Big( 4i \Omega^2(1{-}\Omega^2)^2

2125: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} 

2126: A^\mu(z_1) (\partial_\mu A_\nu)(z_1)

2127: +16\Omega^4 \tilde{z}_1^\mu  A_\mu(z_1)\,

2128: A^\nu(z_1)\Big) \frac{\partial}{\partial

2129:   w^\rho}\Big)\nonumber

2130: \\

2131: & \times e^{-\frac{\tilde{\Omega}(\tau_1 + \tau_2)}{4 \tau_1 \tau_2 (1 +

2132: \Omega^2)T}(1+\tau_1\tau_2)|z_2|^2+4\tau_1\tau_2 \langle

2133: z_2,z_1\rangle +4\tau_1\tau_2 |z_1|^2)

2134: +\frac{2i\Omega^2(\tau_1^2 - \tau_2^2)}{

2135: \tau_1 \tau_2 (1 + \Omega^2)^2T} \langle z_2,\Theta^{-1} z_1\rangle

2136: +\langle w,z_2\rangle}\Big|_{w=0}

2137: \nonumber

2138: \\

2139:  &= \Big(\frac{\tilde{\Omega}(1-\tau_1^2)(1-\tau_2^2)}{

2140: 4\pi(1+\Omega^2)^2 ( \tau_1+\tau_2)(1+\tau_1\tau_2)}\Big)^2

2141: \int d^4 z_1 \; 

2142: e^{-\frac{\tilde{\Omega}(\tau_1 + \tau_2)}{

2143: (1 + \Omega^2)(1+\tau_1\tau_2) } |z_1|^2}

2144: \nonumber

2145: \\

2146: &

2147: \times \Big( \frac{2\tilde{\Omega}(1-\Omega^2)^2(1+\Omega^2) 

2148: (1-\tau_1\tau_2)}{\tau_1+\tau_2} \Big(

2149: A_\mu(z_1)A^\mu(z_1)

2150: \nonumber

2151: \\

2152: & \qquad + A^\mu(z_1)(\partial_\nu A_\mu)(z_1)

2153: \Big(-\frac{2\tau_1\tau_2}{1+\tau_1\tau_2} z_1^\nu

2154: +\frac{i\theta \Omega (\tau_1-\tau_2)}{(1+\Omega^2)(1+\tau_1\tau_2)}

2155: \tilde{z}_1^\nu\Big)

2156: \nonumber

2157: \\ &\qquad 

2158: + 

2159: \frac{A_\mu(z_1)(\partial_\rho\partial_\sigma  A^\mu)(z_1)}{

2160: 2(1+\tau_1\tau_2)^2} 

2161: \Big(2\tau_1\tau_2 z_1^\rho

2162: -\frac{i\theta \Omega (\tau_1-\tau_2)}{(1+\Omega^2)}

2163: \tilde{z}_1^\rho\Big)

2164: \Big(2\tau_1\tau_2 z_1^\sigma

2165: -\frac{i\theta \Omega (\tau_1-\tau_2)}{(1+\Omega^2)}

2166: \tilde{z}_1^\sigma\Big)

2167: \nonumber

2168: \\

2169: &\qquad 

2170: + A^\mu(z_1)(\partial^\nu\partial_\nu  A_\mu)(z_1)

2171: \frac{\tau_1 \tau_2 (1 +\Omega^2)T}{

2172: \tilde{\Omega}(\tau_1 + \tau_2)(1+\tau_1\tau_2)} \Big)

2173: \nonumber

2174: \\

2175: & +16\Omega^4 \tilde{z}_1^\mu  A_\mu(z_1)\,

2176: \tilde{z}_1^\nu A_\nu(z_1)

2177: + (1-\Omega^2)^4 \frac{(\tau_1-\tau_2)^2}{(\tau_1+\tau_2)^2} 

2178: (\partial_\nu A_\mu)(z_1)\,(\partial_\mu A_\nu)(z_1)

2179: \nonumber

2180: \\

2181: &+  \Big( 4i \Omega^2(1-\Omega^2)^2

2182: \frac{\tau_1-\tau_2}{\tau_1+\tau_2} 

2183: A^\mu(z_1) (\partial_\mu A_\nu)(z_1)

2184: +16\Omega^4 \tilde{z}_1^\mu  A_\mu(z_1)\,

2185: A_\nu(z_1)\Big) 

2186: \nonumber

2187: \\

2188: &\qquad\times \Big(

2189: -\frac{2\tau_1\tau_2}{1+\tau_1\tau_2} \tilde{z}_1^\nu

2190: -\frac{2i \tilde{\Omega} (\tau_1-\tau_2)}{(1+\Omega^2)(1+\tau_1\tau_2)}

2191: z_1^\nu\Big)\Big)\;.

2192: %\times e^{\big\langle w, 

2193: %\frac{\tau_1 \tau_2 (1 +\Omega^2)T}{

2194: %\tilde{\Omega}(\tau_1 + \tau_2)(1+\tau_1\tau_2)} w

2195: %-\frac{2\tau_1\tau_2}{1+\tau_1\tau_2} z_1

2196: %+\frac{2i\theta \Omega (\tau_1-\tau_2)}{(1+\Omega^2)(1+\tau_1\tau_2)}

2197: %\Theta^{-1} z_1\big\rangle }\Big|_{w=0}

2198: \label{V2A0}

2199: \end{align}

2200: 

2201: 

2202: \subsection{Moyal vertex plus Moyal-derivative vertex}

2203: 

2204: This combination is (among others) necessary for a new type of

2205: coupling between scalar field and gauge field. We use as far as possible the

2206: same notation as in the previous calculation. Defining the auxiliary

2207: vector $W=(0,0,w_3,w_4)^t$, we have

2208: \begin{align}

2209:   V_2(A,f)&=\int d^4x_1 \,d^4y_1\, d^4x_2 \,d^4y_2\;

2210:   (e^{-(t-t_2)H_0})(y_2,x_1) (L_\star(f))(x_1,y_1)

2211: \nonumber

2212:   \\

2213:   & \times (e^{-t_2H_0})(y_1,x_2) 

2214: \big\{ L_\star(A^\mu) , i \partial_\mu +

2215:   \Omega^2 M_\bullet(\tilde{x}_\mu) \big\}(x_1,y_1) 

2216:   \nonumber

2217:   \\

2218: &=

2219:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2220: 16\pi^2\tau_1\tau_2}\Big)^2

2221: \frac{1}{(2\pi\theta)^8}

2222: \int d^{16} X d^8 Z\; 

2223: f(z_1)A_\mu(z_2)

2224: \nonumber

2225: \\

2226: &\times 

2227:  \Big(2 \tilde{z}_2^\mu  -(1-\Omega^2) (\Theta^{-1})^{\mu\rho} 

2228: \frac{\partial}{\partial w_4^\rho}\Big) 

2229: e^{-\frac{\Omega (\tau_1+\tau_2)}{8\theta \tau_1\tau_2} 

2230: X^t Q X 

2231: - \frac{2}{\theta} X^t \sigma Z +2 X^t W}\Big|_{W=0}

2232: \nonumber

2233: \\

2234: &=

2235:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2236: 16\pi^2\tau_1\tau_2}\Big)^2

2237: \Big(\frac{4 \tau_1\tau_2} {\Omega (\tau_1+\tau_2)}\Big)^8

2238: (\det Q)^{-1/2}

2239: \int d^8 Z\; f(z_1)A_\mu(z_2)

2240: \nonumber

2241: \\

2242: &\times 

2243:  \Big(2 \tilde{z}_2^\mu  +\frac{8i\theta\tau_1\tau_2(1-\Omega^2)}{

2244: \Omega(\tau_1+\tau_2)} (\Theta^{-1})^{\mu\rho} 

2245: (Q^{-1} \tilde{Z})_\rho^4\Big) 

2246: e^{-\frac{8\tau_1\tau_2}{\Omega\theta (\tau_1+\tau_2)}

2247: Z(Q^{-1})Z }

2248: \nonumber

2249: \\

2250: &=

2251:  \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2252: 16(1+\Omega^2)^2\pi^2 T \tau_1\tau_2}\Big)^2

2253: \int d^4 z_1 \,d^4z_2\; f(z_1)A_\mu(z_2)

2254: \nonumber

2255: \\

2256: & \times \Big(

2257: -\frac{i\tilde{\Omega}(\tau_1 - \tau_2)}{2\tau_1 \tau_2 T}

2258: \frac{(1-\Omega^2)^2}{(1+\Omega^2)^2} 

2259: \big((z_1^\mu-z_2^\mu)+\tau_1\tau_2(z_1^\mu+z_2^\mu)\big)

2260: + \frac{\Omega^2(\tau_1+\tau_2)^2}{

2261: (1+\Omega^2)\tau_1\tau_2 T} \tilde{z}_2^\mu\Big)

2262: \nonumber

2263: \\

2264: & \times e^{-\frac{\Omega(\tau_1 + \tau_2)}{2 \theta \tau_1 \tau_2 (1 +

2265: \Omega^2)T}(|z_1-z_2|^2+\tau_1\tau_2|z_1+z_2|^2)

2266: -\frac{2\Omega^2(\tau_1^2 - \tau_2^2)}{\theta

2267: \tau_1 \tau_2 (1 + \Omega^2)^2T} z_1\sigma z_2}

2268: \nonumber

2269: \\

2270: &= \Big(\frac{\tilde{\Omega}^2(1-\tau_1^2)(1-\tau_2^2)}{

2271: 16(1+\Omega^2)^2\pi^2 T \tau_1\tau_2}\Big)^2

2272: \int d^4 z_1 \,d^4z_2\; 

2273: \nonumber

2274: \\

2275: & 

2276: \times \Big( -i\frac{(1-\Omega^2)^2}{1+\Omega^2}

2277: \frac{\tau_1-\tau_2}{\tau_1+\tau_2}  (\partial^\mu f)(z_1)

2278: \,A_\mu(z_2) + \frac{4\Omega^2}{1+\Omega^2} \;

2279: f(z_1)\, \tilde{z}_2^\mu A_\mu(z_2)\Big)

2280: \nonumber

2281: \\

2282: & \times e^{-\frac{\Omega(\tau_1 + \tau_2)}{2 \theta \tau_1 \tau_2 (1 +

2283: \Omega^2)T}(|z_1-z_2|^2+\tau_1\tau_2|z_1+z_2|^2)

2284: -\frac{2\Omega^2(\tau_1^2 - \tau_2^2)}{\theta

2285: \tau_1 \tau_2 (1 + \Omega^2)^2T} z_1\sigma z_2}\;.

2286: \end{align}

2287: As before, for $\tau_1,\tau_2 \to 0$ the integrand is regular unless

2288: $z_1=z_2$, so that we expand 

2289: $A_\mu(z_2)=A_\mu(z_1)+(z_2^\nu-z_1^\nu)\int d\xi \;(\partial_\nu

2290: A_\mu)(z_1+\xi(z_2-z_1))$

2291: and consider the leading term $A_\mu(z_1)$. After a shift $z_2\mapsto

2292: z_2+z_1$ we have, neglecting the subleading summand $\tilde{z}_2^\mu$,

2293: \begin{align}

2294: V_2(A,f)^0 &=  \Big(\frac{\tilde{\Omega}(1-\tau_1^2)(1-\tau_2^2)}{

2295: 4\pi(1+\Omega^2) (\tau_1+\tau_2)(1+\tau_1\tau_2)}\Big)^2

2296: \int d^4 z_1 \; 

2297: e^{-\frac{\tilde{\Omega}(\tau_1 + \tau_2)}{(1 +

2298: \Omega^2)(1+\tau_1\tau_2)}|z_1|^2}

2299: \nonumber

2300: \\

2301: & 

2302: \times \Big( i\frac{(1-\Omega^2)^2}{1+\Omega^2}

2303: \frac{\tau_1-\tau_2}{\tau_1+\tau_2}  f(z_1)

2304: \,(\partial^\mu A_\mu)(z_1) + \frac{4\Omega^2}{1+\Omega^2} \;

2305: f(z_1)\, \tilde{z}_1^\mu A_\mu(z_1)\Big)\;.

2306: \end{align}

2307: 

2308: 

2309: 

2310: \begin{thebibliography}{99}

2311: 

2312: 

2313: %\cite{Grosse:2004yu}

2314: \bibitem{Grosse:2004yu}

2315:   H.~Grosse and R.~Wulkenhaar,

2316: ``Renormalisation of $\phi^4$-theory on noncommutative $\mathbb{R}^4$ 

2317: in the matrix base,''

2318:   Commun.\ Math.\ Phys.\  {\bf 256} (2005) 305

2319:   [arXiv:hep-th/0401128].

2320:   %%CITATION = CMPHA,256,305;%%

2321: 

2322: %\cite{Grosse:2003aj}

2323: \bibitem{Grosse:2003aj}

2324:   H.~Grosse and R.~Wulkenhaar,

2325: ``Power-counting theorem for non-local matrix models and renormalisation,''

2326:   Commun.\ Math.\ Phys.\  {\bf 254} (2005) 91

2327:   [arXiv:hep-th/0305066].

2328:   %%CITATION = CMPHA,254,91;%%

2329: 

2330: 

2331: %\cite{Langmann:2002cc}

2332: \bibitem{Langmann:2002cc}

2333:   E.~Langmann and R.~J.~Szabo,

2334: ``Duality in scalar field theory on noncommutative phase spaces,''

2335:   Phys.\ Lett.\  B {\bf 533}, 168 (2002)

2336:   [arXiv:hep-th/0202039].

2337:   %%CITATION = PHLTA,B533,168;%%

2338: 

2339: %\cite{Grosse:2005da}

2340: %%\cite{Grosse:2004ik}

2341: \bibitem{Grosse:2005da} 

2342: H.~Grosse and R.~Wulkenhaar, ``Renormalisation

2343:   of $\phi^4$-theory on non-commutative $\mathbb{R}^4$ to all

2344:   orders,'' Lett.\ Math.\ Phys.\ {\bf 71} (2005) 13

2345:   [arXiv:hep-th/0403232].

2346:   %%CITATION = LMPHD,71,13;%%

2347:   %%CITATION = HEP-TH/0403232;%%

2348: 

2349: %\cite{Rivasseau:2005bh}

2350: \bibitem{Rivasseau:2005bh}

2351:   V.~Rivasseau, F.~Vignes-Tourneret and R.~Wulkenhaar,

2352: ``Renormalization of noncommutative $\phi^4$-theory by multi-scale  analysis,''

2353:   Commun.\ Math.\ Phys.\  {\bf 262} (2006) 565

2354:   [arXiv:hep-th/0501036].

2355:   %%CITATION = CMPHA,262,565;%%

2356: 

2357: %\cite{Gurau:2005gd}

2358: \bibitem{Gurau:2005gd}

2359:   R.~Gurau, J.~Magnen, V.~Rivasseau and F.~Vignes-Tourneret,

2360: ``Renormalization of non-commutative $\phi^4_4$ field theory in $x$-space,''

2361:   Commun.\ Math.\ Phys.\  {\bf 267} (2006) 515

2362:   [arXiv:hep-th/0512271].

2363:   %%CITATION = CMPHA,267,515;%%

2364: 

2365: %\cite{Gurau:2006yc}

2366: \bibitem{Gurau:2006yc}

2367:   R.~Gurau and V.~Rivasseau,

2368: ``Parametric representation of noncommutative field theory,''

2369:   Commun.\ Math.\ Phys.\  {\bf 272} (2007) 811

2370:   [arXiv:math-ph/0606030].

2371:   %%CITATION = CMPHA,272,811;%%

2372: 

2373: %\cite{Rivasseau:2007qx}

2374: \bibitem{Rivasseau:2007qx}

2375:   V.~Rivasseau and A.~Tanasa,

2376: ``Parametric representation of `critical' noncommutative QFT models,''

2377:   arXiv:math-ph/0701034.

2378:   %%CITATION = MATH-PH/0701034;%%

2379: 

2380: %\cite{Rivasseau:2007fr}

2381: \bibitem{Rivasseau:2007fr}

2382:   V.~Rivasseau,

2383: ``Constructive Matrix Theory,''

2384:   arXiv:0706.1224 [hep-th].

2385:   %%CITATION = ARXIV:0706.1224;%%

2386: 

2387: %\cite{Magnen:2007uy}

2388: \bibitem{Magnen:2007uy}

2389:   J.~Magnen and V.~Rivasseau,

2390: ``Constructive $\phi^4$ field theory without tears,''

2391:   arXiv:0706.2457 [math-ph].

2392:   %%CITATION = ARXIV:0706.2457;%%

2393: 

2394: %\cite{Grosse:2004by}

2395: \bibitem{Grosse:2004by}

2396:   H.~Grosse and R.~Wulkenhaar,

2397: ``The beta-function in duality-covariant noncommutative $\phi^4$-theory,''

2398:   Eur.\ Phys.\ J.\  C {\bf 35} (2004) 277

2399:   [arXiv:hep-th/0402093].

2400:   %%CITATION = EPHJA,C35,277;%%

2401: 

2402: %\cite{Disertori:2006uy}

2403: \bibitem{Disertori:2006uy}

2404:   M.~Disertori and V.~Rivasseau,

2405: ``Two and three loops beta function of non commutative $\phi^4_4$ theory,''

2406:   Eur.\ Phys.\ J.\  C {\bf 50} (2007) 661

2407:   [arXiv:hep-th/0610224].

2408:   %%CITATION = EPHJA,C50,661;%%

2409: 

2410: %\cite{Disertori:2006nq}

2411: \bibitem{Disertori:2006nq}

2412:   M.~Disertori, R.~Gurau, J.~Magnen and V.~Rivasseau,

2413: ``Vanishing of beta function of non commutative $\phi^4_4$ theory to all

2414: orders,''

2415:   Phys.\ Lett.\  B {\bf 649} (2007) 95

2416:   [arXiv:hep-th/0612251].

2417:   %%CITATION = PHLTA,B649,95;%%

2418: 

2419: %\cite{Rivasseau:2007ab}

2420: \bibitem{Rivasseau:2007ab}

2421:   V.~Rivasseau,

2422: ``Non-commutative renormalization,''

2423:   arXiv:0705.0705 [hep-th].

2424:   %%CITATION = ARXIV:0705.0705;%%

2425: 

2426: %\cite{Wulkenhaar:2006si}

2427: \bibitem{Wulkenhaar:2006si}

2428:   R.~Wulkenhaar,

2429: ``Field Theories On Deformed Spaces,''

2430:   J.\ Geom.\ Phys.\  {\bf 56} (2006) 108.

2431:   %%CITATION = JGPHE,56,108;%%

2432: 

2433: %\cite{Langmann:2003cg}

2434: \bibitem{Langmann:2003cg}

2435:   E.~Langmann, R.~J.~Szabo and K.~Zarembo,

2436: ``Exact solution of noncommutative field theory in background magnetic

2437: fields,''

2438:   Phys.\ Lett.\  B {\bf 569} (2003) 95

2439:   [arXiv:hep-th/0303082].

2440:   %%CITATION = PHLTA,B569,95;%%

2441: 

2442: %\cite{Langmann:2003if}

2443: \bibitem{Langmann:2003if}

2444:   E.~Langmann, R.~J.~Szabo and K.~Zarembo,

2445: ``Exact solution of quantum field theory on noncommutative phase spaces,''

2446:   JHEP {\bf 0401} (2004) 017

2447:   [arXiv:hep-th/0308043].

2448:   %%CITATION = JHEPA,0401,017;%%

2449: 

2450: %\cite{Grosse:2005ig}

2451: \bibitem{Grosse:2005ig}

2452:   H.~Grosse and H.~Steinacker,

2453: ``Renormalization of the noncommutative $\phi^3$-model through the  Kontsevich

2454: model,''

2455:   Nucl.\ Phys.\  B {\bf 746} (2006) 202

2456:   [arXiv:hep-th/0512203].

2457:   %%CITATION = NUPHA,B746,202;%%

2458: 

2459: %\cite{Grosse:2006tc}

2460: \bibitem{Grosse:2006tc}

2461:   H.~Grosse and H.~Steinacker,

2462: ``Exact renormalization of a noncommutative $\phi^3$ model in 6 dimensions,''

2463:   arXiv:hep-th/0607235.

2464:   %%CITATION = HEP-TH/0607235;%%

2465: 

2466: %\cite{Grosse:2006qv}

2467: \bibitem{Grosse:2006qv}

2468:   H.~Grosse and H.~Steinacker,

2469: ``A nontrivial solvable noncommutative $\phi^3$ model in 4 dimensions,''

2470:   JHEP {\bf 0608} (2006) 008

2471:   [arXiv:hep-th/0603052].

2472:   %%CITATION = JHEPA,0608,008;%%

2473: 

2474: %\cite{VignesTourneret:2006nb}

2475: \bibitem{VignesTourneret:2006nb}

2476:   F.~Vignes-Tourneret,

2477: ``Renormalization of the orientable non-commutative Gross-Neveu model,''

2478:   Annales Henri Poincar\'e {\bf 8} (2007) 427

2479:   [arXiv:math-ph/0606069].

2480:   %%CITATION = AHPJF,8,427;%%

2481: 

2482: %\cite{VignesTourneret:2006xa}

2483: \bibitem{VignesTourneret:2006xa}

2484:   F.~Vignes-Tourneret,

2485: `` Renormalisation des th\'eories de champs non commutatives,'' Ph.D.\

2486: thesis,   arXiv:math-ph/0612014.

2487:   %%CITATION = MATH-PH/0612014;%%

2488: 

2489: %\cite{Gurau:2005qm}

2490: \bibitem{Gurau:2005qm}

2491:   R.~Gurau, V.~Rivasseau and F.~Vignes-Tourneret,

2492: ``Propagators for noncommutative field theories,''

2493:   Annales Henri Poincar\'e {\bf 7} (2006) 1601

2494:   [arXiv:hep-th/0512071].

2495:   %%CITATION = AHPJF,7,1601;%%

2496: 

2497: %\cite{Chepelev:2000hm}

2498: \bibitem{Chepelev:2000hm}

2499:   I.~Chepelev and R.~Roiban,

2500: ``Convergence theorem for non-commutative Feynman graphs and

2501: renormalization,''

2502:   JHEP {\bf 0103} (2001) 001

2503:   [arXiv:hep-th/0008090].

2504:   %%CITATION = JHEPA,0103,001;%%

2505: 

2506: %\cite{Lakhoua:2007ra}

2507: \bibitem{Lakhoua:2007ra}

2508:   A.~Lakhoua, F.~Vignes-Tourneret and J.~C.~Wallet,

2509: ``One-loop beta functions for the orientable non-commutative Gross-Neveu

2510: model,''

2511:   arXiv:hep-th/0701170.

2512:   %%CITATION = HEP-TH/0701170;%%

2513: 

2514: %\cite{Matusis:2000jf}

2515: \bibitem{Matusis:2000jf}

2516:   A.~Matusis, L.~Susskind and N.~Toumbas,

2517: ``The IR/UV connection in the non-commutative gauge theories,''

2518:   JHEP {\bf 0012} (2000) 002

2519:   [arXiv:hep-th/0002075].

2520:   %%CITATION = JHEPA,0012,002;%%

2521: 

2522: %\cite{Connes:1994yd}

2523: \bibitem{Connes:1994yd}

2524: A. Connes, {\it Noncommutative geometry}, Academic Press Inc., San

2525: Diego (1994) \\

2526: available at \url{http://www.alainconnes.org/downloads.html}.

2527: %\href{http://www.slac.stanford.edu/spires/find/hep/www?irn=6658121}{SPIRES

2528: %entry}

2529: 

2530: %\cite{Connes:1996gi}

2531: \bibitem{Connes:1996gi}

2532:   A.~Connes,

2533: ``Gravity coupled with matter and the foundation of non-commutative

2534: geometry,''

2535:   Commun.\ Math.\ Phys.\  {\bf 182} (1996) 155

2536:   [arXiv:hep-th/9603053].

2537:   %%CITATION = CMPHA,182,155;%%

2538: 

2539: %\cite{Chamseddine:1996zu}

2540: \bibitem{Chamseddine:1996zu}

2541:   A.~H.~Chamseddine and A.~Connes,

2542: ``The spectral action principle,''

2543:   Commun.\ Math.\ Phys.\  {\bf 186} (1997) 731

2544:   [arXiv:hep-th/9606001].

2545:   %%CITATION = CMPHA,186,731;%%

2546: 

2547: %\cite{Chamseddine:2006ep}

2548: \bibitem{Chamseddine:2006ep}

2549:   A.~H.~Chamseddine, A.~Connes and M.~Marcolli,

2550: ``Gravity and the standard model with neutrino mixing,''

2551:   arXiv:hep-th/0610241.

2552:   %%CITATION = HEP-TH/0610241;%%

2553: 

2554: %\cite{Gayral:2003dm}

2555: \bibitem{Gayral:2003dm}

2556:   V.~Gayral, J.~M.~Gracia-Bond\'{\i}a, B.~Iochum, T.~Sch\"ucker and 

2557: J.~C.~V\'arilly,

2558: ``Moyal planes are spectral triples,''

2559:   Commun.\ Math.\ Phys.\  {\bf 246} (2004) 569 

2560:   [arXiv:hep-th/0307241].

2561:   %%CITATION = CMPHA,246,569;%%

2562: 

2563: %\cite{Gayral:2004ww}

2564: \bibitem{Gayral:2004ww}

2565:   V.~Gayral and B.~Iochum,

2566: ``The spectral action for Moyal planes,''

2567:   J.\ Math.\ Phys.\  {\bf 46} (2005) 043503 

2568:   [arXiv:hep-th/0402147].

2569:   %%CITATION = JMAPA,46,043503;%%

2570: 

2571: %\cite{Gayral:2004cs}

2572: \bibitem{Gayral:2004cs}

2573:   V.~Gayral,

2574: ``Heat-kernel approach to UV/IR mixing on isospectral deformation

2575: manifolds,''

2576:   Annales Henri Poincar\'e {\bf 6} (2005) 991

2577:   [arXiv:hep-th/0412233].

2578:   %%CITATION = AHPJF,6,991;%%

2579: 

2580: %\cite{Rieffel:1993??}

2581: \bibitem{Rieffel:1993??} 

2582: M.~A.~Rieffel, \emph{Deformation quantization for actions of $\mathbb{R}^d$},

2583: Mem.\ Amer.\ Math.\ Soc.\ {\bf 106}, Providence (1993).

2584: 

2585: %\cite{de Goursac:2007gq}

2586: \bibitem{de Goursac:2007gq}

2587:   A.~de Goursac, J.~C.~Wallet and R.~Wulkenhaar,

2588: ``Noncommutative induced gauge theory,''

2589:   arXiv:hep-th/0703075.

2590:   %%CITATION = HEP-TH/0703075;%%

2591: 

2592: %\cite{Grosse:2007dm}

2593: \bibitem{Grosse:2007dm}

2594:   H.~Grosse and M.~Wohlgenannt,

2595: ``Induced Gauge Theory on a Noncommutative Space,''

2596:   arXiv:hep-th/0703169.

2597:   %%CITATION = HEP-TH/0703169;%%

2598: 

2599: %\cite{Grosse:2006hh}

2600: \bibitem{Grosse:2006hh}

2601:   H.~Grosse and M.~Wohlgenannt,

2602: ``Noncommutative QFT and renormalization,''

2603:   J.\ Phys.\ Conf.\ Ser.\  {\bf 53} (2006) 764

2604:   [arXiv:hep-th/0607208].

2605:   %%CITATION = 00462,53,764;%%

2606: 

2607: %\cite{Wallet:2007em}

2608: \bibitem{Wallet:2007em}

2609:   J.~C.~Wallet,

2610: ``Noncommutative Induced Gauge Theories on Moyal Spaces,''

2611:   arXiv:0708.2471 [hep-th].

2612:   %%CITATION = ARXIV:0708.2471;%%

2613: 

2614: %\cite{Madore:2000en}

2615: \bibitem{Madore:2000en}

2616:   J.~Madore, S.~Schraml, P.~Schupp and J.~Wess,

2617: ``Gauge theory on noncommutative spaces,''

2618:   Eur.\ Phys.\ J.\  C {\bf 16} (2000) 161

2619:   [arXiv:hep-th/0001203].

2620:   %%CITATION = EPHJA,C16,161;%%

2621: 

2622: %\cite{raimar-habil}

2623: \bibitem{raimar-habil}

2624: R.~Wulkenhaar, ``Renormalisation of noncommutative $\phi^4_4$-theory

2625: to all orders,'' habilitation thesis, Vienna University of Technology

2626: (2005). \\ available at

2627: \url{http://www.math.uni-muenster.de/u/raimar/physics/habilitation/} 

2628: 

2629: %\cite{Blaschke:2007vc}

2630: \bibitem{Blaschke:2007vc}

2631:   D.~N.~Blaschke, H.~Grosse and M.~Schweda,

2632: ``Non-commutative U(1) Gauge Theory on $\mathbb{R}^4$ with Oscillator Term,''

2633:   arXiv:0705.4205 [hep-th].

2634:   %%CITATION = ARXIV:0705.4205;%%

2635: 

2636: %\cite{Gayral:2006wu}

2637: \bibitem{Gayral:2006wu}

2638:   V.~Gayral, J.~H.~Jureit, T.~Krajewski and R.~Wulkenhaar,

2639: ``Quantum field theory on projective modules,''

2640:   arXiv:hep-th/0612048.

2641:   %%CITATION = HEP-TH/0612048;%%

2642: 

2643: %\cite{Connes:1990qp}

2644: \bibitem{Connes:1990qp}

2645:   A.~Connes and J.~Lott,

2646: ``Particle models and noncommutative geometry (expanded version),''

2647:   Nucl.\ Phys.\ Proc.\ Suppl.\  {\bf 18B} (1991) 29.

2648:   %%CITATION = NUPHZ,18B,29;%%

2649: 

2650: %\cite{Gayral:2005ih}

2651: \bibitem{Gayral:2005ih}

2652:   V.~Gayral, J.~M.~Gracia-Bond\'{\i}a and F.~Ruiz Ruiz,

2653: ``Position-dependent noncommutative products: Classical construction and

2654: field theory,''

2655:   Nucl.\ Phys.\  B {\bf 727} (2005) 513

2656:   [arXiv:hep-th/0504022].

2657:   %%CITATION = NUPHA,B727,513;%%

2658: 

2659: %\cite{Simon:1979??}

2660: \bibitem{Simon:1979??}

2661:   B.~Simon,

2662: \emph{Functional integration and quantum physics},

2663: Academic Press, New York (1979).

2664: 

2665: %\cite{eom}

2666: \bibitem{eom}

2667: M.\ Hazewinkel (ed.),\emph{Encyclopaedia of Mathematics}, 

2668: Kluwer Academic Publishers, Dordrecht (2001).

2669: \\

2670: available at \url{http://eom.springer.de/}

2671: 

2672: %\cite{Grosse:2003nw}

2673: \bibitem{Grosse:2003nw}

2674:   H.~Grosse and R.~Wulkenhaar,

2675: ``Renormalisation of $\phi^4$-theory on noncommutative $\mathbb{R}^2$ 

2676: in the matrix base,''

2677:   JHEP {\bf 0312} (2003) 019

2678:   [arXiv:hep-th/0307017].

2679:   %%CITATION = JHEPA,0312,019;%%

2680: 

2681: \end{thebibliography}

2682: \end{document}

2683: 

2684: 

2685: %%% Local Variables:

2686: %%% mode: latex

2687: %%% TeX-master: t

2688: %%% End:

2689: 

2690: 

2691: 

2692: