1: \documentclass{JHEP}
2: \usepackage{epsfig}
3: \usepackage{amssymb}
4: \usepackage[active]{srcltx}
5:
6: \setlength{\oddsidemargin}{0.75in}
7: \setlength{\evensidemargin}{0.75in}
8: \setlength{\topmargin}{1in}
9: \setlength{\textwidth}{7.0in}
10: \setlength{\textheight}{8.5in}
11: \renewcommand{\baselinestretch}{1.2}
12: \jot=2mm
13:
14:
15: \newcommand{\bea}{\begin{eqnarray}}
16: \newcommand{\eea}{\end{eqnarray}}
17: \newcommand{\bean}{\begin{eqnarray*}}
18: \newcommand{\eean}{\end{eqnarray*}}
19: \newcommand{\mat}[1]{\left( \matrix{#1} \right)}
20: \newcommand{\tmat}[1]{{\scriptsize \mat{#1}}}
21: \newtheorem{theorem}{\sf THEOREM}
22: \def\thetheorem{\thesection.\arabic{theorem}}
23:
24: \def\IZ{\mathbb{Z}}
25: \def\IR{\mathbb{R}}
26: \def\IQ{\mathbb{Q}}
27: \def\IC{\mathbb{C}}
28: \def\IP{\mathbb{P}}
29:
30:
31: \def\O #1{\overline{#1}}
32: \def\D #1{\dot{#1}}
33: \def\rightaction#1{\stackrel{\rightarrow}{\partial \over \partial #1}}
34: \def\leftaction#1{\stackrel{\leftarrow}{\partial \over \partial #1}}
35:
36: \def\func#1{\mathop{\rm #1}\nolimits}
37: \def\abs#1{\left| #1\right|}
38: \def\ba{\begin{array}}
39: \def\ea{\end{array}}
40: \def\beq{\begin{equation}}
41: \def\eeq{\end{equation}}
42: \def\braket#1{\left\langle #1 \right\rangle}
43: \def\bra#1{\left\langle #1\right|}
44: \def\ket#1{\left| #1\right\rangle}
45: \def\bbar#1{ \overline #1}
46: \def\Tr{\mathop{\rm Tr}}
47: \def\det{\mathop{\rm det}}
48: \newcommand{\fref}[1]{Figure~\ref{#1}}
49: \def\eref#1{(\ref{#1})}
50: \def\bit#1{\begin{array}{c} \\ \\ \end{array} \hspace{#1 cm}}
51: \def\a{{\alpha}}
52: \def\b{{\beta}}
53: \def\d{{\rm d}}
54: \def\da{{\dot{\alpha}}}
55: \def\db{{\dot{\beta}}}
56: \def\th{{\theta}}
57: \def\bth{{\overline{\theta}}}
58: \def\bm{{\overline{m}}}
59: \def\half{{1 \over 2}}
60:
61: \preprint{
62: {\tt hep-th/0307165}}
63: %
64: \title{${\cal N}=\half$ Wess-Zumino model is renormalizable
65: }
66: \author{Ruth Britto and Bo Feng\\
67: ~~~~\\
68: Institute for Advanced Study \\
69: Einstein Drive, Princeton NJ 08540 \\
70: ~~~~~~~~~~~~~~~~\\
71: \email{britto@ias.edu, fengb@ias.edu} }
72: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
73: \abstract{
74: %%%%%%%%
75: The Wess-Zumino model on ${\cal N}=\half$ nonanticommutative superspace, which contains the dimension-6 term $F^3$,
76: is shown to be renormalizable to all orders in perturbation theory, upon adding $F$ and $F^2$ terms to the original Lagrangian. The renormalizability is possible, even with this higher-dimension operator,
77: because the Lagrangian is not hermitian. Such deformed field theories arise naturally in string theory with a graviphoton background.
78: }
79: %\keywords{}}
80: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
81: \begin{document}
82: %
83: %%
84: \section{Introduction}
85:
86: It is commonly understood that for a four-dimensional quantum field theory to be renormalizable, every operator appearing in the Lagrangian must have a mass dimension less than or equal to 4. An important assumption in this statement is that the action is real.
87: However, it was recently observed \cite{newrefs} that
88: in string theory with a constant graviphoton background, the field theory of the string worldsheet suffers a deformation of its superspace coordinates $\th^\a$. These coordinates no longer anticommute, but satisfy a Clifford algebra: $C^{\a\b}=\{\th^\a,\th^\b\}$. Supersymmetry is reduced by half. It follows too that the action is no longer real. In \cite{seiberg}, it was found that this deformation of superspace effectively adds new terms, of dimension greater than 4, to the Wess-Zumino and Yang-Mills Lagrangians.\footnote{This deformation of superspace was considered previously for example in \cite{non-theta}; recent work on ${\cal N}=\half$ supersymmetric field theories has appeared in \cite{recent,twoloop,bfr,ty,luninrey}.}
89: Initially it was expected that these deformed theories were unrenormalizable, but should simply be treated perturbatively. In this letter we defy intuition by demonstrating the renormalizability of the ${\cal N}=\half$ Wess-Zumino model to all orders. Our result generalizes the recent work \cite{twoloop}, which proved renormalizability up to two loops.
90:
91: The classical Lagrangian is
92: \bea \label{Lag}
93: \int\d^4\th~\O\Phi\Phi + \int\d^2\th~({m \over 2}\Phi^2 + {g \over 3}\Phi^3)+\int\d^2\bth~({\O m \over 2}\O\Phi^2 + {\O g \over 3}\O\Phi^3)~~-~{g \over 3}\det(C)F^3,
94: \eea
95: where
96: \bean
97: \Phi(y,\th) &=& A(y) + \sqrt{2}\th\psi(y) + \th\th F(y) \\
98: \O\Phi(\O y,\O\th)&=&\O A(\O y) + \sqrt{2}\O\th\O\psi(\O y) + \O\th\O\th\O F(\O y),
99: \eean
100: and $y$ and $\O y$ are chiral coordinates.
101: The $F^3$ term arising from the deformation of superspace might appear to render the theory unrenormalizable. However, the recent paper \cite{twoloop}
102: showed that up to two loops it is in fact renormalizable, after adding by hand $F$ and $F^2$ terms to the Lagrangian.
103: Specifically, \cite{twoloop} proved the following assertions up to two loops:
104: \begin{itemize}
105: \item Divergences are at most logarithmic.
106: \item Divergent terms have at most one power of $\det(C)$.
107: \item Divergent terms are of the form $F^\ell \O G^k$, where $\O G \equiv
108: \O m \O A + \O g \O A^2$.
109: \item Contractions with $\O G$ are equivalent to contractions with $F$, and thus the counterterms from $F,F^2,F^3$ suffice to renormalize the theory.
110: \end{itemize}
111: In this work we prove the same assertions to all orders. We work in terms of component fields, making use of the two $U(1)$ (pseudo)symmetries of the theory (see \cite{bfr} and the Appendix), as well as Feynman diagram combinatorics, to constrain the form of divergent terms in the effective action.
112:
113: The reasoning of this kind of renormalizability is discussed in detail in \cite{luninrey}.
114: The essential point is that the higher-dimensional term (like $F^3$)
115: acts as a source vertex with no corresponding sink (like
116: $\O F^3$ or $A^3$), so these vertices do not show up in arbitrarily
117: high order diagrams. This point can be seen clearly in our argument below;
118: for example, the $F^3$ term can appear only once in any divergent
119: diagram.
120:
121:
122:
123: \section{At most log divergence and one power of $\det(C)$}\label{sec-c}
124:
125: In this section we use symmetry arguments to show that the effective action has at most logarithmic divergences, and that the divergent terms have at most one power of $\det(C)$.
126:
127: Suppose a general divergent operator appears in the effective action with a coefficient $\lambda$:
128: \bean
129: \Gamma_{\cal O} = \lambda {\cal O},
130: \eean
131: and that $\lambda$ has mass dimension $d$ and charges $q_R = R, q_{\Phi}=S$ under the two global U(1) (pseudo)symmetries, whose action is described in the Appendix.
132: By dimensional analysis
133: and charge considerations, and with an ultraviolet momentum cutoff, we will have
134: \bea\label{lambda}
135: \lambda\sim \Lambda^d g^{x + 4 z-R} \O g^{x}
136: \left( {m \over \Lambda} \right)^y
137: \left( {\O m \over \Lambda} \right)^{ y + 6 z +{S-3R\over 2}}
138: \left( \det(C) \Lambda^2\right)^z
139: \eea
140: from the loop calculation, where $x,y,z$ are nonnegative integers; the nonegative powers of $m,\O m,g,\O g$ and $\det(C)$ come from the propagators and vertices.
141:
142: The most general operator ${\cal O}$ is given by
143: \bean
144: {\cal O} \sim \partial^p A^{\alpha_1} \O A^{\O \alpha_1} F^{\alpha_2}
145: \O F^{\O \alpha_2} \psi^{\alpha_3} \O \psi^{\O \alpha_3}.
146: \eean
147: The differential operators are understood to act on the others in all possible combinations; for our purposes it is enough to count the overall dimension.
148: Because the term $\Gamma_{\cal O}$ must have mass dimension 4 and zero charge, we must have
149: \bean
150: d & = & 4- p - ( \alpha_1+ \O \alpha_1+ 2 \alpha_2+ 2\O \alpha_2
151: +{3\over 2} \alpha_3+{3\over 2}\O \alpha_3)\\
152: R & = & -\alpha_1+ \O \alpha_1+ \alpha_2 -\O \alpha_2 \\
153: S & = & -\alpha_1+ \O \alpha_1-\alpha_2+\O \alpha_2
154: - \alpha_3+\O \alpha_3
155: \eean
156: Thus, the overall power of $\Lambda$ in $\Gamma_{\cal O}$ is
157: \bea
158: P & = & 4-4z-2y - p - 2\alpha_1- 4 \O \alpha_2-\alpha_3-2 \O \alpha_3.
159: \eea
160: The new operator is superficially divergent iff $P\geq 0$. Note that this restricts $z$, which is the power of $\det(C)$, to be 0 or 1. (1) If $z=0$, it is the ordinary Wess-Zumino case with only the familiar
161: wave function renormalization; we will not discuss it here.
162: (2) If
163: $z=1$, then
164: $y=p=\alpha_1=\O \alpha_2=\alpha_3=\O \alpha_3=0$, and
165: there is at most a logarithmic divergence. The possibly divergent operators take the form $\O A^{\O \alpha_1} F^{\alpha_2}$.
166:
167:
168: \section{Finitely many divergent operators}\label{sec-finite}
169:
170: Now we show that there are only finitely many divergent operators, namely
171: \bea
172: \O A^{\O \a_1} F^{\a_2}, \qquad {\rm for}~~\O \a_1 + 2 \a_2 \leq 6~~{\rm and}~~\a_2 \geq 1.
173: \eea
174: Let us examine the Feynman diagrams that generate a given operator.
175: We introduce the following nonnegative integers to count the number of
176: each type of vertex and propagator that appears in the diagram:
177: \bean
178: a_0:~~F^3,~~~~a_1:~~\O A^2 \O F,~~~~b_1:~~\O A\O \psi \O \psi,~~~~
179: a_2:~~A^2 F,~~~~b_2:~~A\psi \psi,
180: \eean
181: \bean
182: & & c_1:~~A\O A,~~~~c_2:~~F \O F,~~~c_3:~~AF,~~~~c_4:~~\O A\O F,\\
183: & & d_1:~~\psi \psi,~~~~d_2:~~\O \psi \O \psi,~~~~d_3:~~\psi\O \psi.~~~~
184: \eean
185: Now, by matching powers of the coupling constants, we can specify the exponents in the previous section: $(x,y,z) =
186: (a_1 + b_1,c_4+d_2,a_0)$. Moreover, restricting to possibly divergent operators, we apply the main result of the previous section to set $c_4=d_2=0$ and $a_0=1$.
187: By matching the appearances of each field among vertices, contractions and external lines, we can derive six conditions on these integers.
188: %\bea
189: %A:~~~&c_1+c_3 = & 2 a_2+b_2 \\
190: %\O A:~~~& c_1+ \O \alpha_1 = & 2 a_1+ b_1\\
191: %F:~~~& c_2+c_3 + \alpha_2 = & 3+a_2 \label{feqn}\\
192: %\O F:~~~&c_2 = & a_1 \\
193: %\psi:~~~&2d_1+d_3 = & 2 b_2\\
194: %\O \psi:~~~& d_3 = & 2 b_1
195: %\eea
196: One consequence of these conditions is that $d_1 = 6-\O \a_1 -2 \a_2 -c_3$. Because $d_1$ is nonnegative, it follows that $\O \a_1 + 2 \a_2 \leq 6$, which is the desired result. The additional condition $\a_2 \geq 1$ follows from the nonrenormalization of the antiholomorphic superpotential \cite{bfr,ty}.
197:
198: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%5
199: \section{Repackaging in terms of $\O G$}
200:
201: The arguments of the previous two sections have proved the renormalizability of the ${\cal N}=\half$ supersymmetric Wess-Zumino model.
202: However, the observation made in \cite{twoloop} up to two loops is a stronger result:
203: {\it divergent terms can be packaged in terms of $F$ and
204: $\O G \equiv \O m \O A + \O g \O A^2$}.
205: We now show that this result applies generally.
206: The divergent terms generated are
207: \bean
208: F, ~~~F^2, ~~~F^3, ~~~F \O G, ~~~F^2 \O G,~~~F \O G^2.
209: \eean
210: We then invoke the beautiful argument of \cite{twoloop}, that a contraction of any field with $\O G$ is equivalent to its contraction with $F$.\footnote{Being able to trade $\O G$ exactly for $F$ is also a consequence of the nonrenormalization of the antiholomorphic superpotential that was shown in \cite{bfr,ty}.}
211: We conclude that the counterterms for $F, F^2$ and $F^3$ suffice to renormalize the theory.
212:
213: To begin, we simplify our task by neglecting diagrams with fermion loops. This is allowed, because there will be a corresponding bosonic loop that cancels the contribution at least partially. Since the leading order is already log divergent, partial cancellation gives at most a finite total contribution. So we are free to set $b_i=d_j=0$.
214:
215: We will classify our divergent 1PI diagrams by powers of $\O A$ (as external lines) relative to a base diagram with no powers of $\O A$.
216: Consider the following operations on diagrams:
217: \begin{itemize}
218: \item Insertion: Where there is a propagator $AF$, the vertex $\O A^2 \O F$ can be inserted, so that there are now two propagators, $A \O A$ and $F \O F$, and an additional external line $\O A$. It is easy to see that
219: the divergence from the momentum integration is not modified.
220: The replacement can be sketched as $\O m \rightarrow \O g \O A$.
221:
222: \item Deletion: Where there is an external line $\O A$, it must lead to a vertex $\O A^2 \O F$, since there are no fermion loops. Because $c_4=0$, there are no propagators $\O A\O F$. Thus this vertex must be contracted with $A$ and $F$, via the propagators $A \O A$ and $F \O F$. The vertex can be cut out of the diagram, and the two propagators replaced by the single propagator $AF$. Again, the divergence is not modified.
223: \end{itemize}
224: These two operations are inverse. By applying deletion repeatedly, every diagram can be reduced to a `base diagram' with only $F$ on the external lines. To see these two operations more clearly, Figure \ref{f:cutpaste}
225: will be helpful.
226:
227:
228: \EPSFIGURE[ht]{cutpaste.eps,width=12cm}
229: { In this figure, we demonstrate how $\O A F^2$ is related to
230: $ F^2$ by the insertion and deletion operations.
231: \label{f:cutpaste}
232: }
233:
234:
235: We apply insertion to build up all possible diagrams with a given base diagram for $F^\ell$.
236: Let $q$ be the number of $AF$ propagators in the base diagram. From the results in the previous section, we know in fact that $q=6-2\ell$. Note in particular that $q$ is even and $q \leq 4$.
237:
238: From this base, there are ${q \choose k} 2^k$ ways to insert $k$ external lines $\O A$---namely ${q \choose k}$ ways to choose the propagators for insertion, and two choices for the external line from each new vertex $\O A^2 \O F$. Moreover, each insertion yields a diagram with a relative factor ${\O g \over \O m}.$
239:
240: Relative to the base diagram including its symmetry factor $S$, the sum of all the contributions with $k \geq 1$ is
241: \bea
242: S F^\ell \sum_{k=1}^q {q \choose k}2^k ({\O g \over \O m})^k \O A^k,
243: \eea
244: which is simply
245: \bean
246: S F^\ell \left[(1 + 2 {\O g \over \O m}\O A)^q -1 \right].
247: \eean
248: Since $q=2(3-\ell)$,
249: \bean
250: (1 + 2 {\O g \over \O m}\O A)^q -1
251: = (1 + 4 {\O g \over \O m^2}\O G)^{(3-\ell)} -1.
252: \eean
253: Thus the sum over contributions from diagrams with $k\geq 1$, with a given base diagram, is indeed a polynomial in $\O G$, namely
254: \bea
255: S F^\ell \sum_{k=1}^{3-l} {3-l \choose k}4^k ({\O g \over \O m^2})^k \O G^k.
256: \eea
257:
258: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
259:
260: \section{No further divergences from $F$ and $F^2$ vertices}
261:
262: To renormalize the theory, we must add the terms $F$ and $F^2$ by hand to the tree-level Lagrangian. We choose to parametrize these terms as
263: \bea
264: \lambda_1 g^3\O m^4 \det(C) F + \lambda_2 g^2 \O m^2
265: \det(C) F^2
266: \eea
267: The advantage of this parametrization is that both
268: $\lambda_1, \lambda_2$ are dimensionless and charge neutral under
269: both $U(1)_R$ and $U(1)_\Phi$.
270:
271: It is easy to see that in this general situation, the symmetry
272: argument given in section two still applies. For example,
273: we only need to add $\lambda_1^{w_1}\lambda_2^{w_2}$ to
274: equation (\ref{lambda}), which does not affect the power counting.
275: We still have following conclusions. (1) A log divergent term has at most one $\det(C)$ insertion.
276: This
277: means a divergent diagram contains at most one of any of the three
278: vertices $F, F^2, F^3$. (2) The only possible divergent operators, with one power of $\det(C)$, are
279: still
280: \bea
281: \O A^{\O \alpha_1} F^{\alpha_2}, \qquad {\rm for}~~
282: \O \a_1 + 2 \a_2 \leq 2 \ell,~~~~\ell=1,2,3
283: \eea
284: where the vertex $F^{\ell}$ is the one appearing in the diagram.
285:
286: %This more general
287: %condition can be traced back to the condition
288: %\bean
289: % c_2+c_3 + \alpha_2 = \ell+a_2,
290: %\eean
291: %which comes from matching the number of $F$ appearances among propagators, vertices and external lines.
292:
293:
294: The crucial point is that with two new terms $F, F^2$ in the
295: tree level Lagrangian, the insertion and deletion operations
296: are not modified at all. The repackaging of all these divergent
297: terms into $\O G$ will go through unmodified.
298: Thus we conclude that the renormalizability proved in the previous
299: sections continues to hold with the terms $F, F^2$ in the
300: tree level Lagrangian.
301:
302:
303: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
304: %========
305: \section*{Acknowledgements}
306:
307: We thank O. Lunin, S. J. Rey, and N. Seiberg for helpful discussions. This work was supported by the NSF grant PHY-0070928.
308: %
309: %
310:
311:
312:
313: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
314:
315: \appendix
316:
317: \section*{Appendix: The two $U(1)$ (pseudo)symmetries}
318:
319: For the ${\cal N}=\half$ Wess-Zumino model, we can identify two global $U(1)$
320: (pseudo)symmetries
321: by treating all coupling parameters, including the nonanticommutativity parameter $C^{\a\b}$, as the lowest components of (anti)chiral
322: superfields. They are $U(1)_\Phi$ flavor symmetry and $U(1)_R$
323: R-symmetry. Charge assignment is given as follows \cite{bfr}.
324:
325: \begin{center}
326: \begin{tabular}{|c|c|c|c|c|c|c|c|} \hline
327: & dim & $U(1)_R$ & $U(1)_\Phi$ & & dim & $U(1)_R$ & $U(1)_\Phi$\\\hline
328: $\th$ & -1/2 & 1 & 0 & $\O \th$ & -1/2 & -1 & 0 \\ \hline
329: $d\th$ & 1/2 & -1 & 0 & $d\O \th$ & 1/2 & 1 & 0 \\\hline
330: $A$ & 1 & 1 & 1 & $\O A$ & 1 & -1 & -1\\\hline
331: $\psi$ & 3/2 & 0 & 1 & $\O \psi$ & 3/2 & 0 & -1 \\\hline
332: $F$ & 2 & -1 & 1 & $\O F$ & 2 & 1 & -1 \\\hline
333: $g$ & 0 & -1 & -3 & $\O g$ & 0 & 1 & 3 \\\hline
334: $m$ & 1 & 0 & -2 & $\O m$ & 1 & 0 & 2 \\\hline
335: $C^{\alpha \beta}$ & -1 & 2 & 0 & $|C|$ & -2 & 4 & 0 \\\hline \hline
336: $\Phi$ & 1 & 1 & 1 & $\O \Phi$ & 1 & -1 & -1 \\\hline
337: \end{tabular}
338: \end{center}
339:
340:
341: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
342:
343:
344: \bibliographystyle{JHEP}
345: \begin{thebibliography}{99}
346:
347: \bibitem{newrefs}
348: H.~Ooguri and C.~Vafa,
349: %{\em The C-deformation of gluino and non-planar diagrams},
350: arXiv:hep-th/0302109;
351: %%CITATION = HEP-TH 0302109;%%
352: H.~Ooguri and C.~Vafa,
353: %{\em Gravity induced C-deformation},
354: arXiv:hep-th/0303063.
355: %%CITATION = HEP-TH 0303063;%%
356: J.~de Boer, P.~A.~Grassi and P.~van Nieuwenhuizen,
357: %{\em Non-commutative superspace from string theory},
358: arXiv:hep-th/0302078.
359: %%CITATION = HEP-TH 0302078;%%
360:
361: \bibitem{seiberg}
362: N.~Seiberg,
363: %{\em Noncommutative superspace, N = 1/2 supersymmetry, field theory and string theory,}
364: JHEP {\bf 0306}, 010 (2003)
365: [arXiv:hep-th/0305248].
366: %%CITATION = HEP-TH 0305248;%%
367:
368: \bibitem{non-theta}
369: %\cite{Klemm:2001yu}
370: %\bibitem{Klemm:2001yu}
371: S.~Ferrara and M.~A.~Lledo,
372: %{\em Some aspects of deformations of supersymmetric field theories,}
373: JHEP {\bf 0005}, 008 (2000)
374: [arXiv:hep-th/0002084].
375: D.~Klemm, S.~Penati and L.~Tamassia,
376: %{\em Non(anti)commutative superspace,}
377: Class.\ Quant.\ Grav.\ {\bf 20}, 2905 (2003)
378: [arXiv:hep-th/0104190].
379: %%CITATION = HEP-TH 0104190;%%
380:
381:
382: \bibitem{recent}
383: N.~Berkovits and N.~Seiberg,
384: %{\em Superstrings in graviphoton background and N = 1/2 + 3/2 supersymmetry,}
385: JHEP {\bf 0307}, 010 (2003).
386: %%CITATION = HEP-TH 0306226;%%
387: S.~Ferrara, M.~A.~Lledo and O.~Macia,
388: %{\em Supersymmetry in noncommutative superspaces,}
389: arXiv:hep-th/0307039.\\
390: %%CITATION = HEP-TH 0307039;%%
391: J.~H.~Park,
392: %{\em Superfield theories and dual supermatrix models,}
393: arXiv:hep-th/0307060.\\
394: %%CITATION = HEP-TH 0307060;%%
395: T.~Araki, K.~Ito and A.~Ohtsuka,
396: %{\em Supersymmetric gauge theories on noncommutative superspace,}
397: arXiv:hep-th/0307076.\\
398: %%CITATION = HEP-TH 0307076;%%
399: R.~Britto, B.~Feng and S.~J.~Rey,
400: %{\em Non(anti)commutative superspace, UV/IR mixing and open Wilson lines,}
401: arXiv:hep-th/0307091.
402: %%CITATION = HEP-TH 0307091;%%
403:
404: \bibitem{twoloop}
405: M.~T.~Grisaru, S.~Penati and A.~Romagnoni,
406: %{\em Two-loop Renormalization for Nonanticommutative N=1/2 Supersymmetric WZ Model,}
407: arXiv:hep-th/0307099.
408: %%CITATION = HEP-TH 0307099;%%
409:
410: %\cite{Britto:2003aj}
411: \bibitem{bfr}
412: R.~Britto, B.~Feng and S.~J.~Rey,
413: %{\em Deformed superspace, N =
414: %1/2 supersymmetry and (non)renormalization theorems},
415: arXiv:hep-th/0306215.
416: %%CITATION = HEP-TH 0306215;%%
417:
418: \bibitem{ty}
419: S.~Terashima and J.~T.~Yee,
420: %{\em Comments on noncommutative superspace,}
421: arXiv:hep-th/0306237.
422: %%CITATION = HEP-TH 0306237;%%
423:
424: \bibitem{luninrey}
425: O.~Lunin and S.~J.~Rey,
426: %{\em Renormalizability of non(anti)commutative gauge theories with N=1/2 supersymmetry,}
427: arXiv:hep-th/03007275.
428:
429: \end{thebibliography}
430:
431: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
432: \end{document}
433:
434: