Method

Efficiency study was made using Mustafa's, Guannan's and Michael's fast data driven simulator package.

10 M \Lambda_c^+'s and 10 M \Lambda_c^-'s were created with flat p_t, rapidity, and centrality distribution. These were susequently decayed, using Pythia 6 (Two decay channels had to be introduced into Pythia: The three body decay channel, and Lambda(1520) with its decays). Afterwards, the p_T and poosition of the daughter particles were smeared using the information from the embedding and 2014 run data. Because we have no proton embedding information about its p_T uncertainty, pion uncertanty was used. Position resolution is not species dependant for sufficiently high p_T.

Cuts used

For the production of \Lambda_c candidates, relatively open cuts were used, and can be tightened and optimized later, when further information is obtained from TMVA or simillar studies of the cuts.

HFT required
Daughter p_T > 0.3 GeV
Hybrid particle ID (TOF information is used if available)
nSigma dE/dx < 3, TOF 1/beta nSigma < 3, TOF p_T > 0.5 GeV
c\tau > 30 um
Daughters DCA < 200 um
cos(theta) > 0.98

Results

Plots of the results are available at:
http://portal.nersc.gov/project/star/simkomir/toyMC.pdf

On the first page control plots of variuos cuts are plotted: Decay length, cos(theta), daughters DCA, and the invariant mass of the corresponding resonance. Tha last may be wider in the case of A+A collisions and can be widened in Pythia (This will have to be checked in data).

The second page features resulting efficiencies of the cuts used, and further control plots: First is a p_T histograms that illustrates the efficiency of the cuts ... the efficiency is shown for the afforementioned cuts. These are always used in addition to the previous ones to show how the efficiency drops (the p_T cuts shows the efficiency of all of the above cuts plus the p_T cuts for the daughters). The next plots show cos(theta) vs p_T, efficiency of the cuts in centrality and the decay length after the DCA K to pi cut.

The simulation above was used in addition with the wrong charge sign triplets to see the dependence of the ratios of signal vs background with various more stringent cuts imposed. Moreover, a more detailed study with real p_T spectra of the Lambda_c and with realistic scaling of N_colls vs centrality is being produced.

To do

Use real p_T spectra and centrality (position resolution depends on centrality)
Produce an estimate of the significance (most likely will be ready by Wednesday)
Analyze rest of the currently available statistics of the run14
Compare with real data and optimize cuts using TMVA