%================================================================
% LaTeX file with preferred layout for the contributed papers to
% the ICHEP Conference 98 in Vancouver
% process with:  latex hep98.tex
%                dvips -D600 hep98
%================================================================
\documentclass[12pt]{article}
\usepackage{epsfig}
\usepackage{amsmath}
\usepackage{hhline}
\usepackage{amssymb}
\usepackage{times}
\usepackage{cite} % GF for [1-4,7] instead of [1,3,4,7,2] in citations

\renewcommand{\topfraction}{1.0}
\renewcommand{\bottomfraction}{1.0}
\renewcommand{\textfraction}{0.0}
\newlength{\dinwidth}
\newlength{\dinmargin}
\setlength{\dinwidth}{21.0cm}
\textheight23.5cm \textwidth16.0cm
\setlength{\dinmargin}{\dinwidth}
\setlength{\unitlength}{1mm}
\addtolength{\dinmargin}{-\textwidth}
\setlength{\dinmargin}{0.5\dinmargin}
\oddsidemargin -1.0in
\addtolength{\oddsidemargin}{\dinmargin}
\setlength{\evensidemargin}{\oddsidemargin}
\setlength{\marginparwidth}{0.9\dinmargin}
\marginparsep 8pt \marginparpush 5pt
\topmargin -42pt
\headheight 12pt
\headsep 30pt \footskip 24pt
\parskip 3mm plus 2mm minus 2mm

\graphicspath{{psfiles/}}

%===============================title page=============================

% Some useful tex commands
%
\newcommand{\GeV}{\mbox{Ge\hspace{-0.08em}V}}
\newcommand{\MeV}{\mbox{Me\hspace{-0.08em}V}}
%\newcommand{\pb}{\rm pb}
%\newcommand{\cm}{\rm cm}
%\newcommand{\hdick}{\noalign{\hrule height1.4pt}}

\newcommand{\ie}{i.\,e.\ }
\newcommand{\grad}{^\circ\!}
\newcommand{\dstar}{$D^*$}
\newcommand{\dstarmath}{D^*}
\newcommand{\dstarmathp}{D^{*+}}
\newcommand{\dstarpm}{$D^{*\pm}$}
\newcommand{\dstarpjmath}{\dstarmath\! \mbox{+jet}}
\newcommand{\dstarpj}{$\dstarpjmath$}
\newcommand{\dstarmjmath}{\dstarmath\! \mbox{-jet}}
\newcommand{\dstarmj}{$\dstarmjmath$}
\newcommand{\dphidsjmath}{\Delta \phi(\dstarmath \! , \mbox{jet})}
\newcommand{\dphidsj}{$\dphidsjmath$}
%\newcommand{\xgdsjmath}{x_\gamma(\dstarpjmath )}
%\newcommand{\xgdsj}{$\xgdsjmath$}
\newcommand{\Ptjmath}{p_t(\mbox{jet} )}
\newcommand{\Ptj}{$\Ptjmath$}
\newcommand{\Etajmath}{\eta(\mbox{jet} )}
\newcommand{\Ptdsmath}{p_t(\dstarmath )}
\newcommand{\Etadsmath}{\eta(\dstarmath )}

\newcommand{\dedx}{$dE\!/\!dx$}

\newcommand{\Mycaption}[1]{\caption{\em #1}}
\newcommand{\Figabclabel}[1]{\bf #1)} % for internal labels of figures 

%\newcommand{\FIXME}[1]{{\tt FIXME: #1}}

\begin{document}

\pagestyle{empty}
\begin{titlepage}

\noindent
\begin{center}
%{\it {\large version of \today}} \\[.3em] 
\begin{small}
\begin{tabular}{llrr}
%Submitted to & \multicolumn{3}{r}{\footnotesize Electronic Access: {\it http://www-h1.desy.de/h1/www/publications/conf/conf\_list.html}} \\[.2em] \hline 
%Submitted to & \multicolumn{3}{r}{\footnotesize {\it www-h1.desy.de/h1/www/publications/conf/conf\_list.html}} \\[.2em] \hline 
Submitted to & & &
\epsfig{file=/h1/www/images/H1logo_bw_small.epsi
,width=2.cm} \\[.2em] \hline
\multicolumn{4}{l}{{\bf
                XXII International Symposium on Lepton-Photon Interactions at High Energy},
                June~30,~2005,~Uppsala} \\
                 & Abstract:        & {\bf 406}    &\\
                 & Sessions: & {\bf Flavor physics} and {\bf QCD and hadron structure}   &\\
\hline
\end{tabular}
\end{small}
\end{center}
\vspace*{2cm}


\begin{center}
  \Large
  {\bf \boldmath
    Photoproduction of Events\\ Containing a \dstarpm\ Meson
    and a Jet \\
    at HERA
    }

  \vspace*{1cm}
    {\Large H1 Collaboration} 
\end{center}

\begin{abstract}

\noindent
Cross sections are measured for photoproduction events containing a \dstarpm\ meson and a jet.
The data were taken with the H1 detector at the $ep$ collider HERA during the years 1999
and 2000.
Photoproduction is selected by detecting the scattered positron at small angles. The cross
sections are determined in the kinematic region $171 < W < 256$~\GeV, $Q^2 < 0.01$~\GeV$^2$,
$p_t(\dstarmath) > 2.0$~\GeV\ and $|\eta(\dstarmath)| < 1.5$, where $W$ is the photon-proton
centre-of-mass energy, $Q^2$ is the photon virtuality and $p_t$ and $\eta$ denote
transverse momentum and pseudorapidity, respectively. Jets are defined by the inclusive
$k_t$~algorithm. The jet with the highest transverse momentum $p_t(jet) > 3$~\GeV\ that does
not contain the \dstar\ meson is taken. Differential cross sections are shown as function 
of $p_t(jet)$, $p_t(\dstarmath)$, $\eta(jet)$, $\eta(\dstarmath)$ and of the azimuthal
differences between
the \dstar\ and the jet, \dphidsj.
The results are compared with perturbative QCD predictions in collinear and $k_t$-factorisation.
%\begin{center}
%  \FIXME{\bf Version 2.2 (\today)}
%\end{center}
\end{abstract}


\end{titlepage}

\pagestyle{plain}

\section{Introduction}
The measurement of inclusive \dstar\ photoproduction at HERA is interesting as a general test
of the charm production mechanism in high energy $ep$ collisions.
The measurements~\cite{H1:DstarGluonDens99,ZEUS:dijetDstar99,H1:dstargammap03_prel,ZEUS:dstargammap02} are consistent with the assumption that heavy quarks are
predominantly produced via a photon-gluon fusion mechanism shown in figure~\ref{fig:feynman}. 

The measurement of a \dstarpj\ pair gives the possibility to investigate further details 
of the charm production mechanism. The assumption is that the
\dstar\ meson and the jet approximate the kinematics of a charm or anti-charm quark and
another parton, be it the other (anti-)charm, a light quark or a gluon.

Apart from single differential cross sections as a function of the transverse momenta and
pseudorapidity of the  \dstar\ and the jet, measurements of
observables of the \dstarpj\ system provide further stringent tests on the QCD calculations. The
measurement of the difference in azimuthal angle \dphidsj\ of the \dstarpj\ system is presented
here. 
In photoproduction events the photon and the proton collide head-on. In collinear
factorisation this applies also for the partons participating in the hard interaction.
Therefore in a $2\to 2$ process the azimuthal angle $\Delta\phi$ between the outgoing
partons in the plane perpendicular to the incoming beams is exactly $180\grad$.
Smaller $\Delta\phi$ can be due to additional (gluon) radiation and/or due to 
intrinsic transverse momenta $k_t$ of the partons initiating the heavy quark production process.
The measurement of $d\sigma\!/\!d$\dphidsj\   provides thus a crucial test of the QCD calculation and
the treatment of higher order QCD contributions. An illustrating sketch of the different
contributions to the \dphidsj\ distribution is shown in figure~\ref{fig:radKtFragEffects}.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Data Selection}
Data of the years 1999 and 2000 with an integrated luminosity of $\mathcal{L} = 51.1$~pb$^{-1}$
are analysed where protons of $E_p = 920$~\GeV\ have been
collided with positrons of $E_e = 27.6$~\GeV.
The data selection is similar to the preliminary inclusive \dstar\ 
measurement~\cite{H1:dstargammap03_prel}.
Tracks of charged particles are measured in the two Central Jet
Chambers~\cite{H1:Detektor97TrackCalMuon} placed concentrically around the beam line in a 
homogeneous magnetic field of 1.15~Tesla and covering the polar angular range\footnote{
  The polar angle $\theta$ is measured with respect to the direction of the colliding protons. 
  } 
$20\grad < \theta < 160\grad$.
The \dstar\ meson is reconstructed via the decay channel\footnote{
  Charge conjugate states are always implicitly included.
} 
$\dstarmathp \to D^0\pi_s^+ \to K^-\pi^+\pi_s^+$. 

Photoproduction is selected through the reconstruction of the scattered positron
in a crystal \v{C}erenkov calorimeter close to the beam line 33~m away from the nominal
interaction point, the electron tagger~\cite{H1:Detektor97}.
The small scattering angle $\pi-\theta_{e'} < 5$~mrad leads to the kinematic range
of $Q^2< 0.01$~\GeV$^2$ and $0.29 < y < 0.65$ where the inelasticity $y$ is reconstructed 
from the reconstructed positron energy $E_{e'}$, $y = 1 - E_{e'}/E_e$.
The range in the inelasticity $y$ corresponds to photon-proton
centre-of-mass energies of $171 < W < 256$~\GeV.
Compared to~\cite{H1:dstargammap03_prel} the transverse momentum cuts for the
\dstar\ and the kaon are lowered.
The following criteria are applied to ensure good \dstar\ reconstruction:
\begin{itemize}
\item transverse momentum requirements for the \dstar\ and its decay tracks
  of $p_t(\dstarmath) > 2.0$~\GeV, $p_t(K,\pi) > 0.3$~\GeV\ and $p_t(\pi_s) > 0.12$~\GeV,
\item the application of particle identification using the specific energy loss \dedx ,
\item a cut on the sum of the transverse momenta of the kaon and the pion from the $D^0$ decay,
   $p_t(K) + p_t(\pi) > 2.2$ \GeV
\item and the cut $p_t(\dstarmath) / S_{E_t} > 0.10$, where
  $S_{E_t}$ is the scalar sum of all reconstructed hadronic-final-state (HFS) objects with polar angle $\theta > 10\grad$, 
  $S_{E_t} = \sum_{HFS}^{\theta > 10 \grad}\sin{\theta_i}E_i$.
\end{itemize}
Details can be found in~\cite{diss:geroflucke05}.

Jets are defined by the inclusive $k_t$-algorithm~\cite{jetKT93}
in the $p_t$-re\-combina\-tion and $\Delta R$-dis\-tance scheme.
The input of the jet algorithm are hadronic-final-state objects (HFS) 
from combining tracks reconstructed in the Central Jet Chamber and clusters in the
calorimeters.
The algorithm is run in the laboratory frame. 
%Since the algorithm is invariant 
%under longitudinal Lorentz boosts, in photoproduction the result is the same as
%if the algorithm would be applied in the photon-proton centre-of-mass system.

To ensure that all decay particles of the \dstar\ candidate end up in the same jet,
the HFS objects
of the \dstar\ decay tracks are replaced by the \dstar\ candidate itself,
\ie the four-vectors that belong to the \dstar\ decay tracks are replaced by
the sum of the momenta of the tracks assigning the kaon and pion mass, respectively.
In events which contain more than one \dstar\ candidate, the jet algorithm is run separately
for each candidate. For each candidate only its own decay tracks are replaced by the 
\dstar\ candidate. 

The minimal required transverse momentum is $\Ptjmath > 3$~\GeV.
To ensure a satisfying jet reconstruction even at these low values
of $p_t$, jets are restricted to the central detector region $|\Etajmath | < 1.5$
where well measured tracks dominate the HFS objects.

The aim is to tag a second object besides the \dstar\ from the hard process.
Therefore jets \emph{not} containing the \dstar\ meson are selected.
For about $50\%$ of the inclusive \dstar\ mesons at least one (non-\dstar -)jet is found,
for about $10\%$ also a second such jet.
If more than one (non-\dstar -)jet is found,
the jet with the highest transverse momentum is chosen.
In total a signal of $588\pm 46$ \dstarpj\ combinations is observed.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{QCD Calculations}
The results will be compared with two leading order calculations supplemented 
with parton showers, the collinear factorising PYTHIA~6.15~\cite{pythia61} and
the $k_t$-factorising
CASCADE~1.2~\cite{cascade,cascadeII,CASCADE12:manual}, as well as collinear
next-to-leading (NLO) calculations in the massive
(FMNR~\cite{Frixione:totHQ95,Frixione:diffHQ95})
and massless (ZMVFNS~\cite{KniehlHein:D*+jet04,Kniehl:massless02}) scheme.
In all calculations the charm mass is set to \mbox{$m_c = 1.5$~\GeV} and 
the fragmentation fraction $f(c\to\dstarmath) = 0.235$~\cite{Gladilin:c->D*99} is applied.

For PYTHIA three different processes are generated separately and added afterwards:
direct photon-gluon fusion, resolved processes with a light quark or a gluon from
the photon and charm excitation processes where a charm quark from the photon
enters the hard scatter. In the latter process the charm quark is treated as a
massless parton like the light quarks whereas in the other processes the charm
mass is accounted for in all steps of the calculation.
Proton and photon density parameterisations CTEQ5L~\cite{CTEQ5:99} and
GRV-H LO~\cite{GRV:92} are used.
Lund string fragmentation is applied as implemented in PYTHIA. For the longitudinal
fragmentation of the charm quark into the \dstar\ meson the Peterson
parametrisation~\cite{Peterson83} is used with $\epsilon_{pet} = 0.078$.
No uncertainties are determined for the PYTHIA prediction.

For the CASCADE calculations the renormalisation scale $\mu_r^2 = 4 m_c^2 + p_t^2$ 
is chosen and the gluon density set A0~\cite{unintGlu_Jung:04} is used. 
The fragmentation is treated as in PYTHIA.
The importance of higher order terms is estimated
by using the sets A0+ and A0- instead which apply the renormalisation scale 
multiplied and divided by two, respectively.
Separately the charm mass has been varied from the default value $m_c = 1.5$~\GeV\  to
$m_c = 1.3$ and $m_c = 1.7$~\GeV.
The largest deviations of these four calculations from the default set define the upper
and lower bound of the uncertainties of the CASCADE calculation. In most regions of phase
space considered in this
analysis the mass variations have the largest effect.


The main parameters of the (massive) FMNR and (massless) ZMVFNS calculations are listed
in the tables~\ref{tab:FMNR} and~\ref{tab:ZMVFNS}.
The massless scheme is expected to work well~\cite{heavyQuarkConcept:01} 
for $p_t \gg m_c$ which is not always satisfied for the phase space considered here.
On the other hand the massive scheme neglects terms $\propto \alpha_s \ln(p_t/m_c)$ and thus
is expected not to be  
be accurate for high transverse momenta.

In the FMNR calculation the charm fragmentation is treated as longitudinal down scaling
of the charm momentum according to the Peterson fragmentation function.
The parameter $\epsilon_{pet} = 0.035$ is chosen~\cite{Nason:99}.
The uncertainty of the cross section prediction
has been estimated as follows: For each bin of the measurement the maximal deviation up- and 
downward obtained by individually varying the renormalisation scale, the factorisation
scale, the charm mass and $\epsilon_{pet}$ as stated in table~\ref{tab:FMNR} 
define the upper and lower bound of the prediction.

\begin{sloppypar}
  ZMVFNS applies fragmentation functions determined in~\cite{BKKfrag:98}.
  The uncertainty of the ZMVFNS calculation is estimated by considering the influence of
  varying the scales simultaneously in opposite directions as stated in table~\ref{tab:ZMVFNS}.
\end{sloppypar}

For both NLO calculations corrections are applied to account for the transition from 
parton level to hadron level jets. The corrections are calculated using PYTHIA.
In PYTHIA parton level jets are constructed from the generated quarks and gluons after
the parton showering step. The \dstarmj\ is identified as the jet containing the
(anti-)charm quark which fragments into the \dstar.
The ratio of hadron and parton level cross sections in each bin is applied
as hadronisation correction factor to the NLO calculations.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Cross Sections Determination}
The bin averaged differential cross section in a variable $Y$ is calculated according to
\begin{equation}
\label{eq:diffcrosssec}
\frac{d\sigma_{vis}(ep \to e' \dstarpjmath X)}{dY} = \frac{N(\dstarmath) \cdot (1-r)}
  {\Delta Y \cdot \mathcal{BR}(\dstarmath \to K \pi \pi_s) \cdot \mathcal{L} \cdot 
    \epsilon }.
\end{equation}
$\Delta Y$ stands for the width of the considered bin and $N(\dstarmath)$ is the number of
\dstar\ mesons associated with a jet in this bin, determined by a fit to the
$\Delta m  = m(K\pi\pi_s) - m(K\pi)$ distributions of the selected \dstar\ candidates.
The correction for reflections from other $D^0$ decays in the
$\Delta m$ distributions is denoted 
$(1-r)=0.965$~\cite{H1:DstarGluonDens99,diss:stefanhengstmann00}, 
$\mathcal{BR}(\dstarmath \to K \pi \pi_s) = 0.0257$~\cite{PDG2004} is the branching ratio of
the analysed 
\dstar\ decay chain, $\mathcal{L}$ % = 51.1$~pb$^{-1}$ 
is the integrated luminosity
and $\epsilon$ is the product of the track acceptance and reconstruction efficiency,
of the efficiency of particle identification and triggering and
of the electron tagger acceptance.
The trigger and track efficiencies as well as the track acceptance have been determined using
a GEANT~3~\cite{geant3} based simulation, the remaining correction factors are deduced from
data.

%The systematic uncertainties arising from the determination of the correction factors
%and of $N(\dstarmath)$ are summarised in table~\ref{tab:systematics}.
%Half of the difference using CASCADE instead of PYTHIA in the determination of the
%factors based on simulations defines the model dependence.

The main systematic uncertainties of the resulting cross sections are due to the uncertainty 
in the track reconstruction efficiency. Further relevant uncertainties arise from the
determination of the electron tagger acceptance, the trigger uncertainty and the model
dependence in the determination of the factors based on simulations.
Quadratically adding the uncertainties of the different sources results in a total systematic
uncertainty of 15-17\% for each measured bin.

%The uncertainty of the calorimeter energy scale is very small since about 60\% of the
%transverse
%momentum of the jets is measured by tracks and about 30\% using the electromagnetic
%energy scale.

\section{Results}
The \dstarpj\ cross section as a function of \Ptj\ and $p_t(\dstarmath)$ are shown in 
figure~\ref{fig:dsJetxsecPt} and compared with predictions of PYTHIA, CASCADE,
FMNR and ZMVFNS.
Both distributions are rapidly falling towards higher transverse momenta 
and
are reasonably well described by all QCD calculations, but the CASCADE prediction shows
a slightly harder spectrum than observed in the data.

Cross sections as a function of the pseudorapidity of the jet and the \dstar ,
respectively, are shown in figure~\ref{fig:dsJetxsecEta}.
They differ noticeably: The
$\Etadsmath$ distribution falls off with increasing values of $\eta$ (similar
to the inclusive analysis) whereas $\Etajmath$ is almost flat.
It has been checked that this difference is not caused by the slightly different kinematic cuts for
the \dstar\ and the jet, as can be seen from the direct PYTHIA distribution.
The comparison of the
$\Etadsmath$  with the $\Etajmath$ cross sections suggests, that the jet cross section contains
not only a charm jet in which case the \dstar\ and the jet distributions would be expected to be
similar. If, however, the jet cross section contains also contributions from hard gluon radiation
from the initial state (such as included in a $\gamma g \to c\bar{c} g$ process), which tend to
populate mainly the forward (large $\eta$) region,  the behaviour of the measured cross section
could be understood. This hypothesis is also supported from comparison with the QCD calculations,
which show exactly this effect. For direct photon processes ($\gamma g \to c\bar{c}$) as simulated
in PYTHIA (\emph{dir.}), the $\eta$ spectra of the jet and the \dstar\ are found to be
similar (figure~\ref{fig:dsJetxsecEta} left),
only after inclusion of the charm excitation processes, which effectively simulate processes
like $\gamma g \to c\bar{c} g$, the $\eta$ spectrum of the jet can be appropriately described.
A similar feature is obtained with $k_t$ factorisation as implemented in CASCADE, but also with
the NLO calculations.
The FMNR calculations describe the slopes reasonably well whereas in ZMVFNS especially
$\Etadsmath$ shows a steeper slope than the data.


In figure~\ref{fig:dsJetxsecOther} the cross section is presented as a function of \dphidsj. 
It is interesting to observe, that only $\sim$ 25\%  
of the measured cross section come from a back-to-back configuration. Such a
configuration would be observed if the dominant process would be $\gamma g \to c\bar{c}$. 
However, a large fraction of events where the \dstar\ and the jet are not back-to-back
suggest a significant contribution
of higher order QCD radiation. 
In the DGLAP type event generator (PYTHIA) such contribution is simulated with the help
of leading log 
parton showers, but also coming from the resolved photon processes and charm excitation. In the
CASCADE calculation the \dphidsj\ distribution is directly correlated with the $k_t$ of the
gluon generating the $c\bar{c}$ pair and therefore sensitive to the un-integrated gluon density.

In the NLO calculations $\dphidsjmath \neq 180\grad$ comes entirely from real gluon emission
and
therefore the calculations are essentially LO for this observable. It is interesting to mention,
that due to infrared sensitivity the NLO calculations (FMNR and ZMVFNS)  give a huge negative
cross section in the highest \dphidsj\ bin which is compensated by a large positive
contribution in the
next lower \dphidsj\ bin. Therefore both bins are averaged to be compared to the measurement.
It is also interesting to observe, that the $\dphidsjmath \approx 100\grad$ region is not
described by 
the FMNR or ZMVFNS calculations indicating the sensitivity to even higher order 
contributions. 
Such higher order contributions could come from one or more partons radiated in the initial
or final state. In CASCADE and PYTHIA these contributions are simulated with parton showers.
It is still open whether resummations to all orders (as implemented in the parton shower
approach) are needed or whether a parton level calculations of ${\cal O}(\alpha_s^3)$
is already sufficient to describe the measurements.

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Conclusions}
Photoproduction of \dstar\ mesons has been analysed at relatively low 
transverse momenta.
Cross sections have been determined for events with a \dstar\ meson and
a jet not containing the \dstar.
The results have been compared with four approaches of QCD calculations.

The measured transverse momentum distributions of the \dstar\ and jet
are in general in agreement with the
prediction within the  theoretical uncertainties. 

Comparing the pseudorapidity distributions of the \dstar\ and the jet it is
striking that the $\Etadsmath$ falls in forward direction whereas $\Etajmath$
is almost constant. 
This observation suggests that the jet cross section  contains
not only a charm jet but also a significant contribution form a further parton, most likely a
gluon jet.
The gluon radiation can be described by several approaches: By parton showers in
combination with charm excitation processes (PYTHIA), an initial gluon ladder 
in $k_t$ factorisation with initial state parton showers (CASCADE) or 
the presence of a gluon radiation in the hard process (NLO calculations).

The azimuthal difference \dphidsj\ is sensitive to the amount of gluon radiation.
The measured cross section shows, that only $\sim 25\%$ come from a LO process 
($\gamma g \to c\bar{c}$) with a collinear gluon. 
The higher orders contribute significantly. The \dphidsj\ is reasonably well described by
calculations applying leading log parton showers in the collinear factorisation ansatz, or by
using un-integrated gluon densities in the $k_t$-factorisation ansatz.
It is interesting to observe, that the smaller \dphidsj\ region   is not described by 
the FMNR or ZMVFNS calculations indicating the sensitivity to even higher order 
contributions. 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\clearpage
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%\bibliographystyle{nextsummer}
%\bibliography{/afs/desy.de/user/f/flucke/h1/tex/makebibtex/database/dstar,/afs/desy.de/user/f/flucke/h1/tex/makebibtex/database/jet,/afs/desy.de/user/f/flucke/h1/tex/makebibtex/database/h1detector}
\begin{thebibliography}{10}
\expandafter\ifx\csname url\endcsname\relax
  \def\url#1{{\tt #1}}\fi
\expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
\providecommand{\eprint}[2][]{\url{#2}}

\bibitem{H1:DstarGluonDens99}
C.~Adloff, {\em et~al.\/} (H1), Nucl. Phys., {\bf B545}, (1999) 21.
  \eprint{hep-ex/9812023}.

\bibitem{ZEUS:dijetDstar99}
J.~Abbiendi, {\em et~al.\/} (ZEUS), Eur. Phys. J., {\bf C6}, (1999) 67.
  \eprint{hep-ex/9807008}.


\bibitem{H1:dstargammap03_prel}
H1, {\em Photoproduction of $D^*$ Mesons at HERA\/}, International
  Euro\-phy\-sics Con\-fe\-rence on High Energy Physics, EPS03, Aachen (2003).
  Abstract 097, \eprint{H1prelim-03-071},
  \newline 
  \urlprefix\url{http://www-h1.desy.de/h1/www/publications/htmlsplit/
    \newline \hspace*{\fill} H1prelim-03-071.long.html}.

\bibitem{ZEUS:dstargammap02}
ZEUS, {\em Measurement of $D^*$ Photoproduction at HERA\/}, 31st International
  Conference on High Energy Physics, ICHEP '02, Amsterdam (2002). Abstract 786,
  \mbox{\eprint{ZEUS-prel-02-004}},
  \newline 
  \urlprefix\url{http://www-zeus.desy.de/physics/phch/conf/
      \newline \hspace*{\fill} amsterdam\_paper.html}.

\bibitem{H1:Detektor97TrackCalMuon}
I.~Abt, {\em et~al.\/} (H1), Nucl. Instrum. Meth., {\bf A386}, (1997) 348.

\bibitem{H1:Detektor97}
I.~Abt, {\em et~al.\/} (H1), Nucl. Instrum. Meth., {\bf A386}, (1997) 310.

\bibitem{diss:geroflucke05}
G.~Flucke, {\em Photoproduction of $D^*$ Mesons and $D^*$ Mesons Associated
  with Jets at HERA\/}, Ph.D. thesis, Universit\"at Hamburg (2005).
  \eprint{DESY-THESIS-2005-006}.

\bibitem{jetKT93}
S.~Ellis, D.~Soper, Phys. Rev., {\bf D48}, (1993) 3160.
  \eprint{hep-ph/9305266}.

\bibitem{pythia61}
T.~Sj\"ostrand, {\em et~al.\/}, Comput. Phys. Commun., {\bf 135}, (2001) 238.
  \eprint{hep-ph/0010017}.

\bibitem{cascade}
H.~Jung, G.~P. Salam, Eur. Phys. J., {\bf C19}, (2001) 351.
  \eprint{hep-ph/0012143}.

\bibitem{cascadeII}
H.~Jung, Comput. Phys. Commun., {\bf 143}, (2002) 100. \eprint{hep-ph/0109102}.

\bibitem{CASCADE12:manual}
H.~Jung, {\em The CCFM Monte Carlo Generator CASCADE Version 1.2007\/} (2004).
  \urlprefix\url{http://www.desy.de/\symbol{126}jung/cascade/cascade12007.ps.g%
z}.

\bibitem{Frixione:totHQ95}
S.~Frixione, M.~L. Mangano, P.~Nason, G.~Ridolfi, Phys. Lett., {\bf B348},
  (1995) 633. \eprint{hep-ph/9412348}.

\bibitem{Frixione:diffHQ95}
S.~Frixione, P.~Nason, G.~Ridolfi, Nucl. Phys., {\bf B454}, (1995) 3.
  \eprint{hep-ph/9506226}.

\bibitem{KniehlHein:D*+jet04}
G.~Heinrich, B.~A. Kniehl, Phys. Rev., {\bf D70}, (2004) 094035.
  \eprint{hep-ph/0409303}.

\bibitem{Kniehl:massless02}
B.~A. Kniehl, {\em in\/} {\em 14th Topical Conference on Hadron Collider
  Physics\/} (eds. M.~Erdmann, T.~M\"uller), 161--170 (Springer, Heidelberg,
  2003). \eprint{hep-ph/0211008}.

\bibitem{Gladilin:c->D*99}
L.~Gladilin, {\em Charm Hadron Production Fractions\/} (1999).
  \eprint{hep-ex/9912064}.

\bibitem{CTEQ5:99}
H.~L. Lai, {\em et~al.\/} (CTEQ), Eur. Phys. J., {\bf C12}, (2000) 375.
  \eprint{hep-ph/9903282}.

\bibitem{GRV:92}
M.~Gl\"uck, E.~Reya, A.~Vogt, Phys. Rev., {\bf D46}, (1992) 1973.

\bibitem{Peterson83}
C.~Peterson, D.~Schlatter, I.~Schmitt, P.~Zerwas, Phys. Rev., {\bf D27}, (1983)
  105.

\bibitem{unintGlu_Jung:04}
H.~Jung, {\em in\/} {\em XII International Workshop on Deep Inelastic
  Scattering (DIS 2004)\/} (eds. D.~Bruncko, J.~Ferencei,
  P.~Str\'{\i}\v{z}enec), vol.~1, 299--302 (SAS, 2004).
  \eprint{hep-ph/0411287}.

\bibitem{heavyQuarkConcept:01}
W.-K. Tung, S.~Kretzer, C.~Schmidt, J. Phys., {\bf G28}, (2002) 983.
  \eprint{hep-ph/0110247}.

\bibitem{Nason:99}
P.~Nason, C.~Oleari, Nucl. Phys., {\bf B565}, (2000) 245.
  \eprint{hep-ph/9903541}.

\bibitem{BKKfrag:98}
J.~Binnewies, B.~A. Kniehl, G.~Kramer, Phys. Rev., {\bf D58}, (1998) 014014.
  \eprint{hep-ph/9712482}.

\bibitem{diss:stefanhengstmann00}
S.~Hengstmann, {\em A Measurement of Diffractive Charm Production at HERA\/},
  Ph.D. thesis, Universit\"at Z\"urich (2000).

\bibitem{PDG2004}
S.~Eidelman, {\em et~al.\/} (Particle Data Group), Phys. Lett., {\bf B592},
  (2004) 1.

\bibitem{geant3}
R.~Brun, {\em et~al.\/}, {\em GEANT 3 User's Guide\/} (1987).
  \eprint{CERN-DD/EE/84-1}.

\end{thebibliography}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%\clearpage
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{table}[hbp]
  \begin{center}
    \begin{tabular}{l|r|r|c|c|c|c}
      FMNR&\multicolumn{1}{|c|}{$\mu_r$}&\multicolumn{1}{|c|}{$\mu_f$}& $m_c$ [\GeV]
      & $\epsilon_{pet}$ & $p$-PDF & $\gamma$-PDF \\ \hline
      decreasing $\sigma$ &$2\cdot m_t$   & $m_t$         & 1.7 & 0.042 & & \\
      default             &$m_t$          & $2\cdot m_t$  & 1.5 & 0.035 & CTEQ5M~\cite{CTEQ5:99} &GRV-G HO~\cite{GRV:92}\\
      increasing $\sigma$ &$0.5 \cdot m_t$& $4 \cdot m_t$ & 1.3 & 0.028 & & \\ \hline
    \end{tabular}
    \Mycaption{Renormalisation ($\mu_r$) and factorisation ($\mu_f$) scales, the charm 
      pole mass ($m_c$), the Peterson parameters ($\epsilon_{pet}$) and the
      parton density parametrisations of the \dstar\ cross section calculations with 
      the FMNR program with their default settings and variations. The transverse mass is 
      defined as $m_t = \sqrt{m^2_c + (p^2_{t,c}+p^2_{t,\bar{c}})}$.
      }
    \label{tab:FMNR}
  \end{center}
\end{table}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{table}[hbp]
  \begin{center}
    \begin{tabular}[htb]{l*{5}{|c}}  
      ZMVFNS& Fragmentation & $\mu_r$ & $\mu_f$ & $p$-PDF & $\gamma$-PDF\\
      \hline
      decreasing $\sigma$ &                         & $2\cdot m_t$ & $1.25\cdot m_t$ &  &\\
      default             & BKK O~\cite{BKKfrag:98} & $m_t$        & $2\cdot m_t$    
      & CTEQ5M~\cite{CTEQ5:99} & GRV~\cite{GRV:92} \\
      increasing $\sigma$ &                         & $ m_t$       & $4\cdot m_t$    &  &   \\
      \hline
    \end{tabular}
    \Mycaption{Main parameters of the \emph{ZMVFNS} next-to-leading order QCD calculation
      with their default settings and variations which lead to an increasing and decreasing
      prediction of the total visible cross section, respectively.
      The treatment of the charm fragmentation into \dstar\ mesons, the factorisation and
      renormalisation scales $\mu_f$ and $\mu_r$ and
      the parton density parametrisations of the proton and the photon are given.
      The \emph{transverse} mass is defined as 
      $m_t = \sqrt{m^2_c + p^2_t(\dstarmath)}$ with the charm mass $m_c = 1.5$~\emph{\GeV.}
      }
      \label{tab:ZMVFNS}
  \end{center}
\end{table}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\clearpage
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{figure}[p]
  \begin{center}
    \begin{tabular}{c|cc}
      direct & \multicolumn{2}{c}{resolved} \\
      & & \rule[-1.ex]{0pt}{3.3ex} {\sl charm excitation} \\
      \includegraphics[width=.3\textwidth]{dirphot.eps} &
      \includegraphics[width=.3\textwidth]{resphot.eps} &
      \includegraphics[width=.3\textwidth]{charmexgluprop2.eps} \\
    \end{tabular}
    \Mycaption{The main charm production mechanisms in leading order photoproduction:
      direct and resolved processes where the latter are dominated by charm excitation in
      the photon.
      }
    \label{fig:feynman}
  \end{center}
\end{figure}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{figure}[p]
  \begin{center}
    \includegraphics[width=.3\textwidth]{2to2processrad} \hfill
    \includegraphics[width=.3\textwidth]{2to2processkt} \hfill
    \includegraphics[width=.3\textwidth]{2to2processfragm}
    \setlength{\unitlength}{\textwidth}
    \begin{picture}(0,0)
      \put(-0.99,0.27){\Figabclabel{a}}
      \put(-0.65,0.27){\Figabclabel{b}}
      \put(-0.3,0.27){\Figabclabel{c}}
    \end{picture}

    \Mycaption{
      View perpendicular to the photon-proton direction,
      illustrating how a radiated gluon with momentum $\vec{p}_g$ (a)
      or transverse momenta $k_t$ of the incoming partons (b)
      can lead to a non back-to-back topology of the two
      outgoing leading hard partons.
      The momenta $\vec{p}_1$ and $\vec{p}_2$ of these partons would be exactly back-to-back in
      a $2\to 2$ process if the incoming
      partons taking part in the hard interaction collide head-on,
      like in the collinear factorisation approach.
      In (c) it is illustrated that the fragmentation of the charm quark with
      momentum $\vec{p}_2$ into a \dstar\ meson with momentum $\vec{p}_{\dstarmath}$ leads only
      to a small deviation \dphidsj\ from $180\grad$ where the
      jet is emerging from the parton with momentum $\vec{p}_1$.\newline
      }
    \label{fig:radKtFragEffects}
  \end{center}
\end{figure}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%\begin{table}[p]
%  \begin{center}
%    \begin{tabular}{|l|c|}
%      \hline
%      \textbf{Source} &{\boldmath $\sigma_{sys} [\%]$}\\ \hline
%      \multicolumn{2}{|c|}{inclusive \dstar } \\ \hline
%      triggering & 5.5 \\
%      $e$-tagger acceptance & 6 \\%\Red{6} \\ %\Red{3 .. 12} \\
%      track efficiency  & 11 \\ %\Red{11}  \\
%      particle ID &   2 \\ %\hline
%      $\mathcal{BR}(\dstarmathp \to K^- \pi^+ \pi_s^+)$ & 2.5 \\
%      other $D^0$ decays & 1.5 \\ 
%      determination of $N(\dstarmath)$ & 3 \\
%      luminosity $\mathcal{L}$ & 1.5 \\ \hline 
%      \multicolumn{2}{|c|}{\dstarpj\ } \\ \hline 
%      model dependence  & $< 9$ ($<6$ except lowest $\Delta\phi$)\\
%      energy scale & $1.5$ \\ 
%      \hline\hline
%      {\bf total} & $\sim 15 - 17$ \\
%      \hline
%    \end{tabular}
    
%    \Mycaption{Summary of the systematic uncertainties of the \dstarpj\ measurement. 
%      The uncertainty sources above the line are the same as for the inclusive \dstar\ 
%      analysis.}
%    \label{tab:systematics}
%  \end{center}
%\end{table}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{figure}[p]
  \begin{center}
    \setlength{\unitlength}{\textwidth}
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig1a.eps} \hfill
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig1b.eps}
    \begin{picture}(0,0)
      \put(-0.85,0.37){\Figabclabel{a}}
      \put(-0.33,0.37){\Figabclabel{b}}
    \end{picture}

    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig1c.eps} \hfill
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig1d.eps}
    \begin{picture}(0,0)
      \put(-0.85,0.37){\Figabclabel{c}}
      \put(-0.33,0.37){\Figabclabel{d}}
    \end{picture}

    \Mycaption{\dstarpj\ cross sections in bins of the transverse momentum of the
      jet and the \dstar\ compared with the predictions of PYTHIA and CASCADE (a,c) and of 
      next-to-leading order calculations (b,d). FMNR is a \emph{massive}
      and ZMVFNS a \emph{massless} calculation.
      For PYTHIA the direct contribution of the prediction
      is shown separately and labelled as \emph{``dir.''}.
      The central FMNR prediction is shown before and after applying the hadronisation
      corrections.
      }
    \label{fig:dsJetxsecPt}
  \end{center}
\end{figure}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{figure}[p]
  \begin{center}
    \setlength{\unitlength}{\textwidth}

    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig2a.eps} \hfill
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig2b.eps}
    \begin{picture}(0,0)                  
      \put(-0.91,0.37){\Figabclabel{a}}   
      \put(-0.33,0.37){\Figabclabel{b}}   
    \end{picture}                         
                                          
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig2c.eps} \hfill
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig2d.eps}
    \begin{picture}(0,0)
      \put(-0.91,0.37){\Figabclabel{c}}
      \put(-0.39,0.37){\Figabclabel{d}}
    \end{picture}


    \Mycaption{\dstarpj\ cross sections in bins of $\Etadsmath$ (a,b) and 
      $\Etajmath$ (c,d)
      compared with the predictions of PYTHIA and CASCADE on the left and of 
      next-to-leading order calculations on the right. FMNR is a \emph{massive}
      and ZMVFNS a \emph{massless} calculation.
      For PYTHIA the direct contribution of the prediction
      is shown separately and labelled as \emph{``dir.''}.
      The central FMNR prediction is shown before and after applying the hadronisation
      corrections.
      }
    \label{fig:dsJetxsecEta}
  \end{center}
\end{figure}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{figure}[p]
  \begin{center}
    \setlength{\unitlength}{\textwidth}

    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig3a.eps} \hfill
    \includegraphics[width=.49\textwidth]{/h1/psfiles/figures/H1prelim-05-073.fig3b.eps}
    \begin{picture}(0,0)
      \put(-0.92,0.08){\Figabclabel{a}}
      \put(-0.40,0.08){\Figabclabel{b}}
    \end{picture}
    \Mycaption{\dstarpj\ cross sections in bins of \dphidsj\ 
      compared with the predictions of PYTHIA and CASCADE on the left and of 
      next-to-leading order calculations on the right. FMNR is a \emph{massive}
      and ZMVFNS a \emph{massless} calculation.
      For PYTHIA the direct contribution of the prediction
      is shown separately and labelled as \emph{``dir.''}.
      The central FMNR prediction is shown before and after applying the hadronisation
      corrections.
      }
    \label{fig:dsJetxsecOther}
  \end{center}
\end{figure}

\end{document}