cdf

INCLUSIVE PP

DIFFERENTIAL CROSS-SECTIONS

 infn

Authors:

Manuel Mussini,   Niccolo' Moggi   and Franco Rimondi

University and INFN, Bologna

for the CDF Collaboration

Abstract

We present a measurement of the inclusive charged particle pT and ΣEt differential cross section in Minimum-Bias proton-antiproton interactions at sqrt(s)=1960 GeV. The correlation of the particle average transverse pT with the particle multiplicity in the event is also reported. Results are based over 506pb-1 of CDF RunII data and corrected to hadron level. These measures give some of the basic features if the inelastic-non-diffractive particle production spectra.




Blessed plots in brief


Data were collected with the Minimum-Bias trigger in low instantaneous luminosity conditions. Efficiency and acceptance of the trigger have been measured by comparison with a 0-bias sample. Vertex and track reconstruction efficiency were computed with a Pythia tuneA run-dependent MC sample. The calorimeter response to the event ΣEt was also measured with MC.Tracks are selected in pT>0.4 GeV. Both tracks and ΣEt in |η|<1.


The track pT differential cross section d3σ/pTdΦdydpT.
This measure extends the spectrum range a factor 15 wider than the previous CDF measure in 1988. The differential cross-section now spans over 11 orders of magnitude.
In the plot below is shown the total uncertainty: statistical (solid line) and systematic (yellow band).
The new data shows how the function used in Run0 to fit the region [0.4 - 10] GeV becames inadequate at higher pT. The dotted line is the Run0 fit. The continuous green line is a fit to the present data in the same limited region. The same power-law function:
f = A ( p0 / pT+p0 ) ^n
cannot be employed for a fit to the whole spectrum.

(click on the image on the right  for a high resolution eps file)
The convolution of all corrections applied is reflected in the overall variation of the raw to the corrected spectrum as shown here. fig2
Here is shown a comparison with Pythia MonteCarlo simulation at hadron level. Pythia was tuned with the so called tune-A with pT(hat)=1.5 GeV.
The lower plot shows the ratio of data to simulation.

Correlation of the average track pT with the event multiplicity (number of primary tracks). RunII measure is compared with RunI and gives no indication of structures at high multiplicities as may be expected in some non-perturbative production models.

In this plot are used also events collected from a special "high-multiplicity" trigger which selects Minimum-Bias interactions with more that 23 tracks converging to the primary vertex and no other requirements.
Statistical and systematic uncertainties are shown in the lower plot.
Same plot compared to some Pythia Min-Bias production tunings. Regardless all the improvements in the comprehension of low-pT production, the models are still unable to reproduce second order quantities such as final state particle correlations.
When multiple parton interactions are suppressed (``no MPI'') an excessive increase of particle momenta is observed. 		fig5

MC with full calorimeter simulation is employed to evaluate the response of the central calorimeter to the event ΣEt.
The raw ΣEt distribution undergoeas a number of corrections. The most important is shown above and requires that the corredted distribution is "unfolded" to correct for smearing among bins. The unfolding factor is computed with a sample of MC re-weighted to best reproduce the data distribution.
The ΣEt cross-section spectrum d3Et/dηdΦdEt.

The (transverse) energy is measured in the central region only as the sum of the Et of each calorimeter tower in |η|<1. This plot shows the fully corrected distribution. Statistical (solid line) and systematic uncertainties (yellow band) are shown below.<1). the plot shown is the fully corrected distribution. Statistical (solid line) and systematical uncertainties are shown below.

Same plot compared to Pythia simulation at hadron level.

The lower plot shows the ratio of data to simulation.

The measure of the sum Et is very sensitive to multiple interactions. For this reason we selected only crossings with only one reconstructed primary vertex and limited the analysis to the very low-luminosity runs. Still, some undetected multiple interactions remain (mostly cases of overlapping vertices). We can evaluate their amount by studying how the average track multiplicity varies with the instantaneous luminosity. The effect of multiple interactions is then corrcted for.





this page by niccolo' moggi (moggi@fnal.gov)
last update: 2008/07/28