1:
2:
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: % Latex File
5: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6: \documentclass[12pt]{article}
7: \usepackage{amssymb,epsfig}
8: %\usepackage{showkeys}
9:
10: \setlength{\oddsidemargin}{0pt}
11: %\setlength{\textwidth}{15.8cm}
12: \setlength{\textwidth}{15.9cm}
13: \setlength{\topmargin}{-0.5in}
14: \setlength{\textheight}{24.5cm}
15: \addtolength{\jot}{5pt}
16: \addtolength{\arraycolsep}{-3pt}
17: \renewcommand{\textfraction}{0}
18: \renewcommand{\baselinestretch}{1.05}
19: %\renewcommand{\theequation}{\arabic{section}.\arabic{equation}}
20: \renewcommand{\thefootnote}{\fnsymbol{footnote}}
21: \newcommand{\ci}[1]{\cite{#1}}
22: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23:
24: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25: \newcommand{\eps}{\epsilon}
26: \newcommand{\al}{\alpha}
27: \newcommand{\bet}{\beta}
28: \newcommand{\gam}{\gamma}
29:
30: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31: %\newcommand{\be}{\begin{equation}}
32: \newcommand{\be}[1]{
33: %\marginpar{\small\textsf{#1}}
34: \begin{equation}\label{#1}}
35: \newcommand{\ee}{\end{equation}}
36: %
37: \newcommand{\ba}[1]{
38: %\marginpar{\small\textsf{#1}}
39: \begin{eqnarray}\label{#1}}
40: \newcommand{\ea}{\end{eqnarray}}
41: %
42: %\newcommand{\ba}{\begin{eqnarray}}
43: %\newcommand{\ea}{\end{eqnarray}}
44: \newcommand{\baa}{\begin{eqnarray*}}
45: \newcommand{\btab}{\begin{tabular}}
46: \newcommand{\etab}{\end{tabular}}
47: \newcommand{\eaa}{\end{eqnarray*}}
48:
49: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
50:
51: \newcommand{\deriv}{\stackrel{\leftrightarrow}{D}\!}
52: \newcommand{\derleft}{\stackrel{\leftarrow}{D}\!}
53: \newcommand{\derright}{\stackrel{\rightarrow}{D}}
54:
55: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
56: %\def \labeltest #1 {\label{#1} \mbox{\# ${#1}$}}
57: \def \labeltest #1 {\label{#1}}
58: \newcommand\re[1]{(\ref{#1})}
59: \def \qqquad {\qquad\quad}
60: \def \qqqquad {\qquad\qquad}
61: %%%%%%%%%%%%%%%%% Macros %%%%%%%%%%%%%%%%%%%%%%%%
62: \newcommand\lr[1]{{\left({#1}\right)}}
63: \newcommand\lrs[1]{{\left[{#1}\right]}}
64: \def\II{\hbox{{1}\kern-.25em\hbox{l}}}
65: \newcommand \widebar [1] {\overline{#1}}
66:
67: \def \Tr {\mbox{Tr\,}}
68: \def \tr {\mbox{tr}}
69: \newcommand \vev [1] {\langle{#1}\rangle}
70: \newcommand \VEV [1] {\left\langle{#1}\right\rangle}
71: \newcommand \rvev [1] {\langle{#1}\rangle}
72: \newcommand \ket [1] {|{#1}\rangle}
73: \newcommand \bra [1] {\langle {#1}|}
74: \newcommand\bin[2]{\left({#1}\atop{#2}\right)}
75: \newcommand \wt {\widetilde}
76: \def \e {\mbox{e}}
77: \def \CO {{\cal O}}
78: \def \CP {{\cal P}}
79: \def \CT {{\cal T}}
80: \def \CM {{\cal M}}
81: \def \CK {{\cal K}}
82: \def \CH {{\cal H}}
83: \def \CI {{\cal I}}
84: \def \CV {{\cal V}}
85:
86: \newcommand{\inclfig}[2]{\mbox{\epsfxsize=#1cm \epsfbox{#2.ps}}}
87: \newcommand{\insertfig}[2]{\mbox{\epsfxsize=#1cm \epsfbox{#2.eps}}}
88:
89:
90: \begin{document}
91:
92: \begin{titlepage}
93: \begin{flushright}
94: \begin{tabular}{l}
95: TPR--00--23\\
96: hep-ph/0012220
97: \end{tabular}
98: \end{flushright}
99: \vskip1cm
100: \begin{center}
101: {\large \bf
102: Hard Exclusive Production of Tensor Mesons
103: \\}
104: \vspace{1cm}
105: {\sc V.M.~Braun {\rm and} N.~Kivel\footnote{on leave of absence from
106: St.Petersburg
107: Nuclear Physics Institute, 188350, Gatchina, Russia }}
108:
109: \vspace*{0.1cm}
110: {\it
111: Institut f\"ur Theoretische Physik, Universit\"at
112: Regensburg, \\ D-93040 Regensburg, Germany
113: }\\[1.cm]
114: \vskip0.8cm
115: {\bf Abstract:\\[10pt]} \parbox[t]{\textwidth}{
116: We point out that hard exclusive production of tensor mesons $f_2(1270)$
117: with helicity $\lambda=\pm 2$ is dominated by the gluon
118: component in the meson wave function and can be used to determine
119: gluon admixture in tensor mesons in a theoretically clean manner.
120: We present a detailed analysis of the tensor meson distribution amplitudes
121: and calculate the transition form factor $\gamma+\gamma^*\to f_2(1270)$
122: for one real and one virtual photon.}
123: \vskip1cm
124:
125: \end{center}
126:
127: \end{titlepage}
128:
129: \newpage
130:
131:
132: {\large\bf 1.}~~Gluon admixture to meson wave functions is interesting for
133: many reasons
134: and has been subject to extensive and somewhat controversial discussion
135: over many years, see e.g. \ci{gluon}.
136: The purpose of this letter is to point out that
137: separation of quark and gluon contributions in hard processes that are
138: dominated by hadron wave functions at small transverse separations
139: has a different meaning compared to the quark model; in a suitably
140: chosen reaction quark contribution can be down compared to the
141: gluon contribution by a power of the momentum transfer.
142:
143: In particular, we consider hard exclusive production of the tensor
144: meson $f_2(1270)$ with the quantum numbers $J^{PC}=2^{++}$, $I^G=0^+$.
145: This state is non-exotic, and, in quark model, can be constructed
146: either from a quark and an antiquark, or from a pair of gluons.
147: To get the spin projection $s=2$ the quark and the antiquark have to be in
148: a P-wave state, while the gluons can be in $S$-wave.
149: On the light-cone, however, contribution of the orbital angular momentum
150: is higher-twist and we will find that the form factor
151: $\gamma+\gamma^*\to f_2(1270)$ corresponding to a pure helicity state
152: $\lambda=\pm2$ is dominated at large photon virtualities
153: by the gluon contribution.
154:
155: {\large\bf 2.}~~Distribution amplitudes of a tensor meson
156: can be constructed in full analogy with those of vector and pseudoscalar
157: mesons, see e.g. \cite{exclusive,ChZh,BBKT,BB99}.
158: We consider matrix elements of bilocal
159: quark-antiquark operators at a light-like separation $z_\mu$, $z^2=0$:
160: \ba{phi}
161: \langle f_2(P,\lambda)|\bar \psi(z)\gamma_\mu \psi(-z)|0\rangle
162: &=&
163: f_q m^2 p_\mu \frac{e^{(\lambda)}_{\bullet\bullet}}
164: {p_{\scriptstyle\bullet}^2}
165: \int\limits_{-1}^1 \! dt\,e^{itp_{\bullet}} \phi_q(t)
166: \nonumber\\
167: &+&f_q m^2 \left[e^{(\lambda)}_{\mu\bullet} - p_\mu
168: \frac{e^{\scriptstyle\bullet\bullet}}{p_{\bullet}}\right]
169: \frac{1}{p_\bullet}
170: \int\limits_{-1}^1 \! dt\,e^{itp_{\bullet}} g_v(t)
171: +{\cal O}(z_\mu)\,,
172: \nonumber\\
173: \langle f_2(P,\lambda)|\bar \psi(z)\gamma_\mu\gamma_5 \psi(-z)|0\rangle
174: &=&
175: -i f_q m^2 \varepsilon^{\mu\nu\alpha\beta} \frac{z^\nu p^\alpha}{p_\bullet}
176: \frac{e^{(\lambda)}_{\beta\bullet}}{p_\bullet}
177: \int\limits_{-1}^1 \! dt\,e^{itp_{\scriptstyle\bullet}} g_a(t)\,,
178: \ea
179: where $\bar \psi \psi = \bar u u+\bar d d$. $P_\mu$ and $m=1270$~MeV
180: are the $f_2$-meson momentum and mass, respectively: $P^2=m^2$.
181: Here and below we use a shorthand notation $p_\bullet = p_\mu z^\mu$, etc.,
182: and define the light-like vector
183: $p_\mu= P_\mu - z_\mu m^2/(2p_\bullet)$,
184: $p^2=0$. The polarization tensor $e^{(\lambda)}_{\alpha\beta}$ is
185: symmetric and traceless, and satisfies the condition
186: $e^{(\lambda)}_{\alpha\beta} P^\beta=0$.
187: Polarization sums can be calculated using
188: \be{polarization}
189: \sum_\lambda e^{(\lambda)}_{\mu\nu}\left(e^{(\lambda)}_{\rho\sigma}\right)^\ast
190: = \frac12 M_{\mu\rho} M_{\nu\sigma}+\frac12 M_{\mu\sigma} M_{\nu\rho}
191: -\frac13 M_{\mu\nu} M_{\rho\sigma}\,,
192: \ee
193: where $M_{\mu\nu} = g_{\mu\nu} - P_\mu P_\nu/m^2$ and
194: the normalization is such that
195: $e^{(\lambda)}_{\mu\nu}\big(e^{(\lambda')}_{\mu\nu}\big)^\ast=
196: \delta_{\lambda\lambda'}$.
197:
198: The distribution amplitude $\phi_q(t) = -\phi_q(-t)$ is the leading twist-2
199: distribution amplitude for the tensor mesons with helicity $\lambda=0$,
200: while $g_v(t)=-g_v(-t)$ and $g_a(t)=g_a(-t)$
201: correspond to the twist-three amplitudes
202: with helicity $\lambda= 1$. The leading twist-2
203: distribution amplitude for $\lambda= 1$ is given by a similar
204: matrix element with the $\sigma_{\mu\bullet}$ matrix in between the
205: quarks; it does not contribute to two-photon processes
206: (if quark masses are neglected) and will not be
207: considered here.
208:
209: Normalization of the distribution amplitudes is chosen to be:
210: \be{norm}
211: \int\limits_{-1}^{1} dt\, t\,\phi_q(t) =
212: \int\limits_{-1}^{1} dt\, t\,g_v(t) = 1\,,\qquad
213: \int\limits_{-1}^{1} dt\, g_a(t)=0\,.
214: \ee
215: The constant $f_q$ is defined as the matrix element of the local operator
216: \be{norm1}
217: \langle f_2(P,\lambda)|\bar \psi\gamma_\mu
218: i\stackrel{\leftrightarrow}{D}_\nu\psi|0\rangle
219: = f_q m^2 e^{(\lambda)}_{\mu\nu}
220: \ee
221: and was estimated using QCD sum rules \cite{AS82}:
222: \be{gf}
223: f_q(\mu = 1~{\rm GeV}) \simeq 56~{\rm MeV}\,.
224: \ee
225: The twist-three distribution amplitudes $g_v(t)$ and $g_a(t)$ receive
226: a Wandzura-Wilczek type contribution that can be expressed in terms of the
227: distribution function $\phi_q$ of the leading twist. A standard calculation
228: (cf. \cite{BBKT}) yields
229: \ba{WW}
230: g_v^{WW}(t) &=& \int\limits_{-1}^t dw\, \frac{\phi_q(w)}{1-w}+
231: \int\limits_{t}^1 dw\, \frac{\phi_q(w)}{1+w}\,,
232: \nonumber\\
233: g_a^{WW}(t) &=& \int\limits_{-1}^t dw\, \frac{\phi_q(w)}{1-w}-
234: \int\limits_{t}^1 dw\, \frac{\phi_q(w)}{1+w}\,.
235: \ea
236: Because of a large mass of the $f_2$-meson, we expect that the twist-3
237: distribution amplitudes are dominated by these ``kinematic'' contributions
238: whereas genuine twist-3 contributions of quark-gluon operators are relatively
239: small.
240:
241: The conformal expansion of the leading-twist distribution amplitude
242: $\phi_q(t)$ goes over the usual set of Gegenbauer polynomials
243: $(1-t^2)\,C^{3/2}_n(t)$ with odd values of $n=1,3,\ldots$ because of the
244: $C$-parity. The asymptotic wave function is, therefore
245: \be{phi-as}
246: \phi_q^{\rm as}(t) = \frac{15}{4} t(1-t^2)
247: \ee
248: and the corresponding expressions for the twist-3 distributions are
249: \be{gav-as}
250: g_v^{\rm as}(t) = \frac52 t^3\,,\qquad g_a^{\rm as}(t) =
251: \frac54 (3t^2-1)\,.
252: \ee
253: In addition, there exist two different leading twist
254: gluon distribution amplitudes
255: \ba{phi-g}
256: \langle f_2(P,\lambda)|
257: S_{\mu\nu}G^a_{\bullet\mu}(z)G^a_{\bullet\nu}(-z)|0\rangle
258: &=&
259: f^T_g e^{(\lambda)}_{\mu\nu} p_\bullet^2 \int\limits_{-1}^1 \! dt\,
260: e^{itp_{\bullet}} \phi_g^T(t) +{\cal O}(z_\mu, z_\nu)\,,
261: \nonumber\\
262: \langle f_2(P,\lambda)|
263: S_{\mu\nu}G^a_{\xi\mu}(z)G^a_{\xi\nu}(-z)|0\rangle
264: &=&
265: f^S_g e^{(\lambda)}_{\mu\nu} \int\limits_{-1}^1 \! dt\,
266: e^{itp_{\bullet}} \phi_g^S(t) +{\cal O}(z_\mu, z_\nu)\, .
267: \ea
268: Here $S_{\mu\nu}$ stands for the symmetrisation in the two indices and removal of the
269: trace:
270: $ S_{\mu\nu}{\cal O}_{\mu\nu} = \frac12\, {\cal O}_{\mu\nu} +
271: \frac12\,{\cal O}_{\nu\mu}
272: -\frac14\, g_{\mu\nu}{\cal O}_{\xi\xi}$.
273: The distribution amplitudes $\phi_g^T(t)$ and $\phi_g^S(t)$ are
274: both symmetric to the interchange
275: of $t\leftrightarrow -t$ and describe the momentum fraction distribution
276: of the two gluons in the $f_2$-meson having the same and the opposite
277: helicity, respectively.
278: The asymptotic distributions at large scales are equal to
279: \be{phi2as}
280: \phi_g^{T,{\rm as}}(t) = \phi_2^{S,{\rm as}}(t)=
281: \frac{15}{16} (1-t^2)^2\,.
282: \ee
283: The constants $f^T_g$ and $f^S_g$ are defined through the matrix element of
284: the local two-gluon operator:
285: \ba{GTnorm}
286: \langle f_2(P,\lambda)|
287: G^a_{\alpha\beta}(0)G^a_{\mu\nu}(0)|0\rangle &=&
288: f^T_g\biggl\{
289: \big[(P_\alpha P_\mu-\frac12 m^2g_{\alpha \mu})\, e^{(\lambda)}_{\beta\nu}-
290: (\alpha\leftrightarrow\beta)\big]- (\mu\leftrightarrow\nu)\biggl\}
291: \nonumber\\
292: &+&\frac12 f^S_g m^2\biggl\{\big[g_{\alpha \mu}\,e^{(\lambda)}_{\beta\nu}
293: - (\alpha\leftrightarrow\beta)\big]- (\mu\leftrightarrow\nu)\biggl\}.
294: \ea
295:
296: The constant $f^T_g$ is renormalized multiplicatively
297: \cite{LFBK85,Eric,JiH98}, while
298: $f^S_g$ is mixed with $f_q$ \cite{Grozin}:
299: \ba{GTscale}
300: f^T_g(Q^2)&=& f^T_g(\mu^2) L^{-1+6N_c/\beta_0}\,,
301: \nonumber\\
302: f_q(Q^2)&=& \frac{n_f}{n_f+4C_F}
303: \biggl(
304: f^S_g(\mu^2)+f_q(\mu^2)
305: \biggl)
306: -L^{\frac23(n_f+4C_F)/\beta_0}
307: \biggl(
308: f^S_g(\mu^2)-\frac{4C_F}{n_f}f_q(\mu^2)
309: \biggl),
310: \nonumber\\
311: f^S_g(Q^2)&=&\frac{4C_F}{n_f+4C_F}
312: \biggl(
313: f^S_g(\mu^2)+f_q(\mu^2)
314: \biggl)+
315: L^{\frac23(n_f+4C_F)/\beta_0}
316: \biggl(
317: f^S_g(\mu^2)-\frac{4C_F}{n_f}f_q(\mu^2)
318: \biggl),
319: \ea
320: where $L=\alpha_s(Q^2)/\alpha_s(\mu^2)$, $C_F=\frac{N_c^2-1}{2N_c}$ and
321: $\beta_0 = 11/3 N_c -2/3 n_f$.
322: Note that the sum $f^S_g+f_q$ is scale-independent.
323: We are not aware of any estimates of the numerical values of $f^T_g$, $f^S_g$.
324: The main goal of this letter is to point out that the constant $f^T_g$
325: can be measured in hard exclusive processes.
326:
327:
328:
329: {\large\bf 3.}~~In the rest of the work we consider one particular hard
330: reaction: the transition form factor
331: $\gamma^*(q)+ \gamma(q')\rightarrow f_2(P)$
332: for one real $q'^2=0$ and one virtual $q^2=-Q^2$ photon.
333: The amplitude of this process is related to the matrix element
334: \be{Tdef}
335: T_{\mu\nu}=i\int d^4x \,e^{-ix(q-q')/2}
336: \langle f_2(P,\lambda)|T\{ j_\mu(x/2)j_\nu(-x/2)\}|0\rangle,
337: \ee
338: where $j_\mu = e_u \bar u \gamma_\mu u + e_d \bar u \gamma_\mu u+\ldots$
339: is the electromagnetic current.
340: The kinematics is illustrated in Fig.~1.
341: %
342: %\input fig1.tex
343: \begin{figure}[t]
344: \begin{center}
345: %\vspace{-1cm}
346: \hspace{0cm}
347: \insertfig{3.5}{kinf2}
348: \end{center}
349: %\vspace{-0.5cm}
350: \caption[dummy]{
351: \small The kinematics of the process
352: $\gamma^*(q)+\gamma(q')\rightarrow f_2(P)$
353: \label{Fig1}
354: }
355: \end{figure}
356: %
357: Neglecting contributions that vanish after the multiplication by photon
358: polarizations, the amplitude $T_{\mu\nu}$ can be parametrized in terms
359: of the three invariant form factors
360: \ba{Tmunu}
361: T_{\mu\nu} &=&
362: -g_{\mu\nu}^\perp
363: e_{\alpha\beta}(q-q')^\alpha (q-q')^\beta \frac{m^2}{(2qq')^2} \,T_0(Q^2)+
364: \nonumber \\&&
365: -g_{\nu\alpha}^\perp e_{\alpha\beta}(q-q')^\beta
366: \biggl(q-q'\frac{q^2}{(qq')}\biggl)_\mu \frac{m^2}{(2qq')^2}\, T_1(Q^2)
367: \nonumber\\&&
368: +\Big(g_{\mu\alpha}^\perp g_{\nu\beta}^\perp+
369: g_{\mu\nu}^\perp\frac{(q-q')^2}{(2qq')^2}
370: (q-q')^\alpha (q-q')^\beta\Big) e^{\alpha\beta} T_2(Q^2)
371: \biggl],
372: \ea
373: where we have introduced the metric tensor $g_{\mu\nu}^\perp$
374: that is transverse to the photon momenta
375: \be{gT}
376: g_{\mu\nu}^\perp=g_{\mu\nu} -
377: \frac1{(qq')}(q_\mu q'_\nu+q_\nu q'_\mu-\frac{q^2}{(qq')}q'_\mu q'_\nu)\,.
378: \ee
379: The form factors $T_0, T_1$ and $T_2$ correspond to the
380: three possible helicity amplitudes
381: \ba{Thelicity}
382: T_0: && \, \gamma^*(\pm 1)+\gamma(\pm 1)\rightarrow f_2(0)\, ,
383: \nonumber \\
384: T_1: &&\, \gamma^*(0)+\gamma(\pm 1)\rightarrow f_2(\mp 1)\, ,
385: \nonumber \\
386: T_2: &&\, \gamma^*(\pm 1)+\gamma(\mp 1)\rightarrow f_2(\pm 2)\, .
387: \ea
388: In the limit $Q^2\rightarrow 0$ only $T_0$ and $T_2$ contribute and
389: their values at $Q^2=0$ determine the two-photon decay width
390: \be{Ggg}
391: \Gamma_{\gamma\gamma}= \frac{(4\pi\alpha)^2}{60\pi m }
392: \Big( |T_0(0)|^2+2|T_2(0)|^2
393: \Big) ,
394: \ee
395: where $\alpha=1/137$ is the fine structure constant.
396:
397: Experimentally it was found that the contribution of the helicity zero
398: state to the two photon width $\Gamma_{\gamma\gamma}$ is very small
399: (see, for example, \ci{MARKII90, JADE90}). Using \cite{MARKII90}
400: %\footnote{Let us for conveniens provide the
401: %estimation of the photon width
402: %from Review of Particle Physics \ci{PDG}:
403: %$\Gamma_{\gamma\gamma}=2.8\pm 0.4 keV$}:
404: \ba{Width}
405: \Gamma_{\gamma\gamma}=3.15\pm 0.04\pm 0.39~{\rm keV}, \qquad
406: \frac{\Gamma_0(f_2\rightarrow \gamma\gamma)}
407: {\Gamma_2(f_2\rightarrow \gamma\gamma)}< 0.05\,\, (90\%\, \mbox{C.L.})
408: \ea
409: we obtain
410: \be{Tnorm}
411: T_2(0)=(212 \pm 20)\, \mbox{MeV}, \qquad T_0(0)< 67\, \mbox{MeV}.
412: \ee
413:
414:
415:
416: Next, we consider the region where the virtuality of
417: one of the photons is large:
418: \be{hard}
419: -q^2\equiv Q^2 \gg m^2,\, \, q'^2=0.
420: \ee
421: In this kinematics, the form factors
422: $\gamma^*(q)+ \gamma(q')\rightarrow f_2(P)$ can be calculated by the light-cone
423: expansion of the T-product of the electromagnetic currents in \re{Tdef}
424: \ba{OPE}
425: i T\{ j_\mu(x)j_\nu(-x)\} &=&
426: \frac{1}{(4\pi)^2 x^4}\sum e^2_q \Bigg\{\Big[
427: \bar\psi(-x)\gamma_\nu\!\not\!x\gamma_\mu \psi(x)-
428: \bar\psi(x)\gamma_\mu\!\not\!x\gamma_\nu \psi(-x)
429: \Big]
430: \\&& -i \frac{\alpha_s}{4\pi }
431: \int_{-1}^1\!\!\! du\!\!\int_{-1}^1\!\!\! dv\, (1+uv)
432: S_{\mu\nu} G_{x\mu}(ux)G_{x\nu}(-vx)\Bigg\}+ \dots\,,
433: \ea
434: where we have shown the contributions that will be relevant for what follows,
435: and $G_{x\mu} = x_\xi G_{\xi\mu}$, etc. The corresponding Feynman diagrams are
436: shown in Fig.~2. Note that we take into account the contribution
437: of the two-gluon operator with helicity $\lambda=2$ but do not consider
438: the gluon contribution with $\lambda=0$ that enters at the same level
439: as the ${\cal O}(\alpha_s)$ corrections to the quark-antiquark operator.
440: %
441: %\input fig2.tex
442: \begin{figure}[t]
443: \begin{center}
444: %\vspace{-1cm}
445: \hspace{0cm}
446: \insertfig{10}{box}
447: \end{center}
448: \vspace{-0.5cm}
449: \caption[dummy]{
450: \small Leading-order contributions to
451: $\gamma^*(q)+ \gamma(q')\rightarrow f_2(P)$ in the large-$Q^2$ limit.
452: \label{Fig2}
453: }
454: \end{figure}
455: %
456: In this paper we calculate the leading-twist contributions to the three
457: form factors (helicity amplitudes) defined in \re{Tmunu} to the leading order
458: in $\alpha_s$. For the most interesting case of the gluon-dominated amplitude
459: $T_2$ we also keep the leading higher-twist correction.
460: A simple calculation gives:
461: \ba{T012}
462: T_0&=&\frac59 f_q \int_{-1}^1\! \frac{tdt}{1-t^2}\,\phi_q (t)\,,
463: \nonumber\\
464: T_1&=&\frac59 f_q \, \int_{-1}^1\! dt \,\ln(1-t^2)\,g_a^{WW}(t)\,,
465: \nonumber\\
466: T_2&=& 2 \frac{\alpha_s}{\pi} \sum e^2_q\, f_g^T\,\int_{-1}^1
467: \,\frac{dt}{1-t^2}\, \phi_g^T(t)+
468: \nonumber \\
469: &+&\frac59 \frac{f_q m^2}{Q^2}\,\int_{-1}^1\! dt\,
470: \ln\left(\frac{1+t}{1-t}\right)\,g_v^{WW}(t)\,,
471: \ea
472: where all distribution amplitudes and the coupling have to be taken
473: at the scale $\mu^2=Q^2$.
474: Note that to ensure the gauge invariance the higher-twist distribution
475: amplitudes must be taken, to our accuracy,
476: in the Wandzura-Wilczek approximation.
477:
478: It is worthwhile to mention that hard exclusive $f_2$-meson
479: production can be viewed as a particular case of the more general
480: hard exclusive two-pion production process considered in \ci{DPT}.
481: Factorisation theorems derived for the two-pion production
482: \cite{Freund} are valid for our case as well, and the
483: results in \re{T012}, with the exception of the
484: Wandzura-Wilzek contribution to $T_2$, can be
485: extracted from the ones existing in the literature \ci{DPT,KMP,KM00}
486: using a correspondence
487: between the two-pion soft matrix elements and the distribution amplitude
488: of a tensor meson \ci{MPol}.
489: In particular, the hard parton subprocesses and hence the coefficient
490: functions are the same in both cases.
491:
492: We expect that distribution amplitudes of the $f_2$-meson are not far from
493: their asymptotic form given in the text. On this assumption, which is
494: certainly acceptable for an estimate, we obtain for large $Q^2$
495: \be{T2as}
496: T_2 \simeq \frac53\, \frac{\alpha_s}{\pi}\, f^T_g +\frac{50}{27}\,
497: f_q\, \frac{m^2}{Q^2}\, ,
498: \ee
499: where we have taken into account $u, d$ and $s$-quark contribution in the
500: quark loop in the box diagram.
501: %
502: %\input fig3.tex
503: \begin{figure}[t]
504: \begin{center}
505: %\vspace{-1cm}
506: \hspace{0cm}
507: \insertfig{8}{gg2pi}
508: \end{center}
509: %\vspace{-0.5cm}
510: \caption[dummy]{
511: \small The two subprocesses contributing to the reaction
512: $ee\rightarrow ee\pi^+\pi^-$ : $\gamma^*\gamma$ scattering (a) and
513: bremsstrahlung (b).
514: \label{Fig3}
515: }
516: \end{figure}
517: %
518: For a realistic value $Q^2\sim 10$~GeV$^2$ the gluon contribution is comparable
519: to the subleading quark contribution if the constants $f_g^T$ and $f_q $
520: are of the same order. It is interesting to note that $T_2(0)$ is much larger
521: than both $T_0(0)$ \re{Tnorm} and $f_q$ \re{gf}. This may indicate that
522: $f_g^T$ is abnormally large.
523:
524: Since $f_2$ decays in two pions with the branching ratio
525: about 95\%, one natural possibility to measure the form factors
526: is via the hard exclusive two-pion production
527: $ee\to ee \pi\pi $, see Fig.~3a, that received a lot of attention
528: recently \cite{DPT,Freund,KMP,KM00,Diehl}. This reaction is in fact observed
529: and constitites a background to the $\gamma^*\gamma\to \pi$ form factor
530: measurements reported by CLEO \cite{Savinov}.
531: Let $k_1$ and $k_2$ be the momenta of the two pions
532: in the final state. The tensor form factor of interest
533: $T_2(Q^2)$ is related to the two-pion helicity amplitude
534: $A_{+-}$ (we use notations of Ref.~\cite{Diehl}, see Eq.~(75) there)
535: in the region $(k_1+k_2)^2\sim m^2$ as
536: \be{twopion}
537: A_{+-} = \varepsilon_+^\mu\varepsilon_-'^\nu (k_1-k_2)_\mu^\perp
538: (k_1-k_2)_\nu^\perp
539: \frac{g_{f_2\pi\pi}}{m}\frac{T_2(Q^2)}{m^2-(k_1+k_2)^2},
540: \ee
541: where $g_{f_2\pi\pi}$ is the corresponding decay constant:
542: \be{fpipi}
543: \langle \pi(k_1)\pi(k_2)|f_2(P,\lambda)\rangle=\frac{g_{f_2\pi\pi}}{m}
544: e_{\alpha\beta}^{(\lambda)}(k_1-k_2)^\alpha(k_1-k_2)^\beta
545: \ee
546: and $\varepsilon_+^\mu,\,\varepsilon_-'^\nu$ denote the transverse
547: polarisation vectors of the virtual and the real photon, respectively.
548: As was discussed in Ref.~\cite{Diehl,KMP}
549: the amplitude $A_{+-}$ can be separated
550: using its nontrivial dependence on the azimuthal angle $\varphi$.
551: Moreover, $A_{+-}$ is symmetric to the interchange of the
552: pion momenta and can be measured by the interference with the
553: (much larger) contribution of the bremsstrahlung
554: process (see Fig.~3b) that is antisymmetric to the interchange
555: of the pion momenta: In the difference of cross sections
556: $\sigma(k_1,k_2)-\sigma(k_2,k_1)$
557: only the interference term survives.
558: The relevant expressions
559: for the physical cross sections have been worked out in \cite{Diehl}.
560:
561: {\large\bf 4.}~~To summarize, in this letter we point out a
562: possibility to determine gluon admixture in tensor mesons using hard
563: exclusive reactions.
564: We construct the basic theoretical formalism and suggest a particular
565: experimental setup where the relevant form factor can be measured.
566: For simplicity, in this work we did not consider radiative corrections.
567: NLO corrections to the coefficient functions have been calculated in
568: \cite{KMP} and NNLO evolution of the corresponding distributions
569: amplitudes can be found in \cite{BelitskyS, BelitskyT}.
570:
571: \subsection*{Acknowledgements}
572: This work was supported in part by the DFG, project Nr. 920585.
573:
574: \begin{thebibliography}{99}
575:
576: \bibitem{gluon}
577: F.~E.~Close,
578: %``Gluonic Hadrons,''
579: Rept.\ Prog.\ Phys.\ {\bf 51} (1988) 833;\\
580: %%CITATION = RPPHA,51,833;%%
581: F.~E.~Close, G.~R.~Farrar and Z.~Li,
582: %``Determining the gluonic content of isoscalar mesons,''
583: Phys.\ Rev.\ {\bf D55} (1997) 5749.
584: %%CITATION = HEP-PH 9610280;%%
585:
586: \bibitem{exclusive} V.L.\ Chernyak and A.R.\ Zhitnitsky, JETP Lett.\
587: {\bf {25}} (1977) 510;
588: Yad.\ Fiz.\ {\bf 31} (1980) 1053;\\
589: A.V.\ Efremov and A.V.\ Radyushkin, Phys.\ Lett.\ B {\bf 94} (1980)
590: 245;
591: Teor.\ Mat.\ Fiz.\ {\bf {42}} (1980) 147;\\
592: G.P.\ Lepage and S.J.\ Brodsky, Phys.\ Lett.\ B {\bf 87} (1979) 359;
593: Phys.\ Rev.\ D {\bf 22} (1980) 2157.
594:
595: \bibitem{ChZh}
596: V.~L.~Chernyak and A.~R.~Zhitnitsky,
597: %``Asymptotic Behavior Of Exclusive Processes In QCD,''
598: Phys.\ Rept.\ {\bf 112} (1984) 173.
599:
600: \bibitem{BBKT}
601: P.~Ball, V.~M.~Braun, Y.~Koike and K.~Tanaka,
602: Nucl.\ Phys.\ {\bf B529} (1998) 323.
603:
604: \bibitem{BB99}
605: P.~Ball and V.~M.~Braun,
606: Nucl.\ Phys.\ {\bf B543} (1999) 201.
607:
608: \bibitem{AS82}
609: T.~M.~Aliev and M.~A.~Shifman,
610: %``QCD Sum Rules And Tensor Mesons. (In Russian),''
611: Sov.\ J.\ Nucl.\ Phys.\ {\bf 36} (1982) 891.
612:
613: \bibitem{LFBK85}
614: A.~P.~Bukhvostov, G.~V.~Frolov, L.~N.~Lipatov and E.~A.~Kuraev,
615: Nucl.\ Phys.\ {\bf B258} (1985) 601.
616:
617: \bibitem{Eric}
618: E.~Sather and C.~Schmidt,
619: Phys.\ Rev.\ {\bf D42} (1990) 1424.
620:
621: \bibitem{JiH98}
622: P.~Hoodbhoy and X.~Ji,
623: Phys.\ Rev.\ {\bf D58} (1998) 054006.
624:
625: \bibitem{Grozin}
626: D.~J.~Gross and F.~Wilczek,
627: %``Asymptotically Free Gauge Theories. I,''
628: Phys.\ Rev.\ {\bf D8} (1973) 3633.
629:
630: \bibitem{MARKII90}
631: J.~Boyer {\it et al.},
632: %``Two Photon Production Of Pion Pairs,''
633: Phys.\ Rev.\ {\bf D42} (1990) 1350.
634:
635: \bibitem{JADE90}
636: T.~Oest {\it et al.} [JADE Collaboration],
637: %``Measurement Of Resonance Productions In The Reactions Gamma Gamma $\to$ Pi0 Pi0 And Gamma Gamma $\to$ Pi0 Eta,''
638: Z.\ Phys.\ {\bf C47} (1990) 343.
639:
640: %\bibitem{PDG}
641: %C. Caso et al, Eur.\ Phys.\ J.\ {\bf C3} (1998) 1.
642:
643: \bibitem{DPT}
644: M.~Diehl, T.~Gousset, B.~Pire and O.~Teryaev,
645: %``Probing partonic structure in gamma* gamma --> pi pi near threshold,''
646: Phys.\ Rev.\ Lett.\ {\bf 81} (1998) 1782.
647:
648: \bibitem{Freund}
649: A.~Freund,
650: %``Proof of factorization for electroproduction of multiple mesons and exclusive gamma* gamma production of multiple hadrons,''
651: Phys.\ Rev.\ {\bf D61} (2000) 074010.
652:
653: \bibitem{KMP}
654: N.~Kivel, L.~Mankiewicz and M.~V.~Polyakov,
655: %``NLO corrections and contribution of a tensor gluon operator to the process gamma* gamma --> pi pi,''
656: Phys.\ Lett.\ {\bf B467} (1999) 263.
657:
658: \bibitem{KM00}
659: N.~Kivel and L.~Mankiewicz,
660: %``Twist-3 contribution to the gamma* gamma --> pi pi amplitude
661: %in the Wandzura-Wilczek approximation,''
662: hep-ph/0010161.
663:
664:
665: \bibitem{MPol}
666: M.~V.~Polyakov,
667: %``Study of two-pion light-cone distribution amplitudes
668: %in the resonance region and at low energies,''
669: Nucl.\ Phys.\ {\bf B555} (1999) 231.
670:
671: \bibitem{Diehl}
672: M.~Diehl, T.~Gousset and B.~Pire,
673: Phys.\ Rev.\ {\bf D62} (2000) 073014.
674:
675: \bibitem{Savinov}
676: J.~Gronberg {\it et al.} [CLEO Collaboration],
677: Phys.\ Rev.\ {\bf D57} (1998) 33.
678:
679:
680: \bibitem{BelitskyS}
681: A.~V.~Belitsky, D.~Muller, L.~Niedermeier and A.~Schafer,
682: Nucl.\ Phys.\ {\bf B546} (1999) 279.
683:
684: \bibitem{BelitskyT}
685: A.~V.~Belitsky, A.~Freund and D.~Muller,
686: Phys.\ Lett.\ {\bf B493} (2000) 341.
687:
688:
689:
690: \end{thebibliography}
691:
692: \end{document}
693:
694: \appendix
695: \renewcommand{\theequation}{\Alph{section}.\arabic{equation}}
696: \setcounter{table}{0}
697: \renewcommand{\thetable}{\Alph{table}}
698:
699: %\section*{Appendices}
700:
701:
702:
703: \end{document}