Assumptions for estimating yields for VTX physics 

(Tony Frawley, last update August 19, 2003)


Introduction

The official PHENIX set of future running scenarios is being developed now. The existing draft is at Bill's PHENIX long term planning page and it contains lots of very useful stuff. In particular, there is a spreadsheet that summarizes expected integrated luminosities under the various scenarios. There is also a link to the latest guidance from CAD on projected delivered luminosity over the next 5 years.

My goal here is to give you a recipe for taking:
 and producing a realistic estimate of the yield obtained.

Integrated Luminosity estimates 

These should be taken from one of Bill's spreadsheet scenarios (see the link above). The spreadsheet has a parameter in cell A1 that sets the optimism index: 1 is the minimum CAD projection, 2 is the maximum CAD projection, 3 is the geometric mean. I agree with Bill that we should use an optimism index of 3, ie. the geometric mean of the minimum and maximum projections.

The integrated luminosities given in the spreadsheet scenarios are PHENIX integrated luminosities. They are obtained by calculating the integrated luminosity delivered by CAD over the specified number of weeks, using Thomas Roser's model of the luminosity evolution over the years and within each run, and then applying some factors that account for various losses at PHENIX. The factors that are already included are:
 Also given for each scenario is the expected pi0 pT reach and expected inclusive Jpsi yield in muon arm north. These are based on measured yields in Run 3, and are fairly realistic except that they do not include the additional reality factors discussed below. For the present discussion, I will assume that you are going to calculate your own yields and acceptances, since you presumably want other kinds of particles than Jpsi.

Converting BBC luminosity to collisions sampled

Au-Au

Collisions sampled = PHENIX integrated luminosity (b-1) * (BBC cross section = 6.9 b * 0.92 = 6.35 b)

p-p

Collisions sampled = PHENIX integrated luminosity (mb-1) * (BBC cross section = 21.8 mb)

Effect of vertex cuts

For a collision diamond with a gaussian distribution and a sigma of 20 cm (Roser estimate for full storage RF fields):

  Z vertex cut  (cm)         Fraction of luminosity retained
      5                                         0.197
    10                                         0.383
    15                                         0.547
    20                                         0.683
    25                                         0.789
    30                                         0.866
    35                                         0.920

If a vertex cut of 10 cm is used, assume that we retain 38% of the luminosity. Compare with 86% for a 30 cm vertex cut. In that case, to get the 10 cm cut PHENIX integrated luminosity use:

        Bill's PHENIX integrated luminosity * (0.38/0.86 = 0.44)

Additional reality factors (dileptons)

Minbias Au-Au:

Loss of reconstruction efficiency              0.5 (muon arms)
                             - or -                           0.33 (central arms)

(use the square root of this number for single leptons in AuAu)

Good run selection                                   0.8

Trigger efficiency                                     0.9 (???)

Au-Au total additional reality factor    0.36 (muon arms)
                   - or -                                     0.24 (central arms)

p-p

Good run selection                                   0.8

Trigger efficiency                                     0.8 (???)


p-p total additional reality factor          0.64

Example

Take the example of  Jpsi yield expected from a 5 weeks setup + 19 week production Au-Au run in 2010 (Run 10), with 10 cm vertex cuts. From Bill's spreadsheet, the expected PHENIX integrated luminosity using optimism index 3 (geometric mean of minimum and maximum projections) is 701 µb-1 (remember that the PHENIX luminosity in Bill's spreadsheet assumes 30 cm vertex cuts).  


Correct the PHENIX integrated luminosity for a 10 cm vertex cut instead of a 30 cm vertex cut.

                        PHENIX integrated luminosity (10 cm cut) = 701 * 0.44  =  308 µb-1
 
Get the expected number of Jpsi's into the (acceptance * single particle reconstruction efficiency) of the detector. This requires either the production cross section for the particle in p-p collisions + the (acceptance * reconstruction efficiency) from simulations, or the measured yield in PHENIX for a known integrated luminosity.  Bill has already estimated the north muon arm Au-Au Jpsi yields from Run 3 p-p Jpsi yields as follows, and shows them in his spreadsheet:


Then we expect into the north muon arm (acceptance * reconstruction efficiency)

    1970.92 * 1970.92 *  1.6 * 10-3 Jpsi per µb-1 * 308 µb-1  = 8213

Now we have to add the dimuon reality factors for loss of reconstruction efficiency, good run selection, and trigger efficiency.

    Final Jpsi estimated yield in north muon arm = 8213 * 0.36 = 2957