Dear Brahmens,
I have tried to extract the ratio
of 2 to 1 neutrons from the ZDC spectrum for events with
nothing in the tiles or beam -beam counters.
I fit the distribution between 40 and 600 GeV
to 9 gaussians whose positions are
fixed at 65GeV, 130GeV and 195GeV etc
and whose width is sigma, sqrt(2)*sigma, sqrt(3)*sigma etc
where sigma is the 10th fit parameter.
I get a ratio of .33 +- .02 +-.02 where the 2nd systematic error
comes from varying the fit range.
Sigma = 25% and when I look at the fit this seems too wide.
If I force the fit to use a lower sigma then the ratio goes down.
The chi**2/NDF is 134/110 which seems to be OK.
I have also tried to add a constant background to the fit
but this lowers the ratio since the 2n peak is wider than the
1 neutron peak.
I have studied the correlation function of the left and right
energy distributions. For the events in my sample the correlation
function is flat as a function of the energy difference in the
two calorimeters. However if I select "nuclear" events by
allowing hits in the tiles I get a strong peak in the
correlation function that is about 150GeV (or 2.5 neutrons)
wide. In other words for nuclear events
the number of neutrons going left and right is correlated but
this is not true for EM events.
From the correlation study I think that I do not have a
significant nuclear contamination in my sample. However my
ratio of .33 +.02 +-.02. disaggrees with Alexie Denisov (of Phenix)
of
0.46 +/ -0.03(stat)+/- 0.05(syst.) .
Alexie has the width of the 2nd neutron peak to be 1.9 * the
width of the first, I think that this is an overestimate.
If it was sqrt(2)*sigma_1n his result would be much closer to
mine.
Yours Michael
Michael Murray, Cyclotron TAMU, 979 845 1411 x 273, Fax 1899
This archive was generated by hypermail 2b29 : Thu Jan 11 2001 - 13:15:58 EST