Photoproduction of Charm

Recent measurements of inclusive photoproduction of $D^*$ mesons were performed by H1 [7] and by ZEUS [19] using the decay channel $D^*\rightarrow D^0\pi_s$ with $D^0 \rightarrow K^-\pi^+ (+c.c)$. The $D^*$ signal used for the ZEUS analysis is shown in fig.11 in section 5. In fig.21 the ZEUS data are compared with predictions from two next-to-leading order calculations, namely the fixed order massive calculation FMNR [80] and the matched calculation FONLL [92] (sections 2.1 and 3.2.1). For the calculations the scale parameters are chosen to be $\mu_R=\mu_F=m_T=\sqrt{m_c^2+p_{c,T}^2}$, where $\mu_R$ is the renormalization scale parameter and $\mu_F$ is the factorization scale parameter, and $m_c=1.5$ GeV. The uncertainties are estimated by variation of $\mu_R$ from $0.5 m_T$ to $2 m_T$ and $m_c$ from $1.3$ GeV to $1.7$ GeV. Good general agreement is seen with relatively large theoretical uncertainties as estimated by simultaneous variation of the renormalization scale and the mass of the charm quark. The central values of the NLO predictions reproduce the shape of the d$\sigma/dW$ distribution and general trends of the $d\sigma /dp_t$ distributions. However, the central NLO predictions significantly underestimate the data over almost the whole kinematic range. The FONLL predictions do not provide a better description of the data than does the NLO calculation. For large $p_t(D^*)$, the FONLL predictions lie further below the data than does the NLO calculation.

In fig.22 $D^*$ photoproduction data from H1 are shown [7]. In this data sample the low angle scattered electron is detected in an electron detector situated 33 m away from the interaction point close to the beam pipe in electron direction. This requirement, necessary for trigger purposes, leads to a constraint on the range in $W_{\gamma p}$ of the data sample to $171<W_{\gamma p}<256$ GeV and restricts the statistical precision. The range in $Q^2$ is restricted to $<0.01$ GeV$^2$. The data are compared with the fixed order massive calculation from [80] and the massless calculation from [96]. For the massive calculation the renormalization and factorization scales were chosen as $2 \mu_R = \mu_F = 2 \sqrt{m_c^2+p_{t,c}^2}$, different from the choice of ZEUS (see above). For calculation in the 4-flavour massless scheme the BKK fragmentation function has been applied [96,98] and the renormalization and factorization scales have been chosen as $\mu_R = \mu_F = 2 \sqrt{m_c^2+p_{t,c}^2}$ for the central prediction. The theories agree in general with the data. However, the massive calculation appears to produce somewhat too hard a $p_t$ spectrum while the massless calculation fits the data better both in shape and in normalization.

Figure: Differential cross sections for inclusive $D^*$ photoproduction in the process $ep\rightarrow D^* X$ :a) $d\sigma /dp_t$, b) $d\sigma /d\eta$, c) $d\sigma /dW$ and d) $d\sigma /dz$ [19]. The predictions of NLO calculations with the central choice of parameters are given by the solid histograms [80]. The dashed histograms show the scale uncertainties of the fixed order NLO prediction (see text). The FONLL predictions [92], are shown as dotted curves and their uncertainties are given by the shaded bands.

Figure: Differential $D^*$ photoproduction cross section $d\sigma /dp_t$ as a function of $p_t$ of the $D^*$ meson. The data are compared with NLO QCD calculations in the 3-flavour massive [80] and in the 4-flavour massless scheme [96] (taken from [7]).