Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://zebu.uoregon.edu/~uochep/talks/talks04/prog07.pdf Äàòà èçìåíåíèÿ: Fri Jul 16 02:44:13 2004 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 10:22:56 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: m 80

Progress rep ort on hh
Olya Igonkina (University of Oregon)
July 7, 2004

This talk concentrates on non-resonant hh, states only. (14 channels) · Pre-selection, SP5/SP6 comparison · PID, backgrounds · Selection, cuts optimization · MC luminocity scaling · Expected sensitivity
hh O.Igonkina

= e, µ, h = , K

1


Preselection (similar to

The pre-selection is based on Tau1N skim, with additional cuts on

· Event has either L3OutDch or L3OutEmc trigger bit set · Event has BGFMultiHadron filter bit set · 3 nGoodTrkLoose 5 · neLoose + nmuLoose 1 (data only) · Exactly 4 `good tracks' are required in the event. (GTVL, no conversions) · The total charge of the good tracks is equal to 0. · 1-3 topology according to thrust hemisphere

hh O.Igonkina

2


Comparison of Run1-3/SP5 and Run4/SP6
Sample - e- K + K - [%] data [%] b¯[%] b cc[%] ¯ uds[%] [%] di-muon [%] Bhabha [%] - e- K + K data [%] b¯[%] b cc[%] ¯ uds[%] [%] di-muon [%]
-

Trig 94.4 56.6 95.6 89.0 83.5 37.6 1.01 19.9 94.2 57.1 96.2 89.8 83.9 38.3 0.94

BGFMH 87.5 97.7 99.4 97.7 95.1 39 15.8 2.28 87.7 94.5 99.3 97.6 95 39.3 16.4

[%]

Tau1N 4 tracks RUN 1-3 / SP5 55.7 88.2 100 28.5 1.03 34.9 4.94 43.2 7.31 51 60 75.6 15.8 6.3 27.4 6.37 RUN 4 / SP6 56.3 88.7 100 29 0.878 35.1 4.43 43.2 6.74 50.8 59 74.5 16.1 5.74

Qi = 0 98.7 86.3 65.7 75.8 83.1 98.6 79.4 73.4 98.7 85.8 66.4 76.4 83.6 98.5 77.7

1-3 topology 99.9 98.1 94.9 96.8 97.4 99.9 99.2 99.7 99.9 98 94.7 96.5 97.1 99.9 99.4

Total 42.4 13.3 0.214 1.36 2.39 6.56 0.00125 0.0058 43.2 13.2 0.185 1.24 2.22 6.49 0.0011

No significant difference was found between SP5 and SP6 MC (signal and backgrounds). Therefore all data were combined together.
hh O.Igonkina 3


Particle identification
All standard PID selectors were checked for best performance. The suppression of background (estimated as a PID-weight of lumi-scaled MC events) to few events in the signal region was required. The PID selectors must be common to all related channels. The final list is PidLHElectrons, muNNTight, KLHTight and piLHLoose.

- - - - - - - - - - - - - -



e- K + K - e- K + - e- + K - e- + - µ- K + K - µ- K + - µ- + K - µ- + - e+ K - K - e+ K - - e+ - - µ+ K - K - µ+ K - - µ+ - -

Signal 0.551 0.623 0.646 0.730 0.300 0.387 0.394 0.475 0.532 0.637 0.731 0.291 0.383 0.467

b¯ b 0.012 0.046 0.034 0.098 0.008 0.029 0.018 0.075 1.5 · 10-4 0.025 0.079 1.5 · 10-4 0.014 0.088

cc ¯ 0.007 0.030 0.002 0.009 0.006 0.031 0.003 0.012 3.0 · 10-4 0.003 0.002 2.1 · 10-4 0.004 0.007

uds 3.9 · 10-4 4.3 · 10-4 4.1 · 10-4 0.002 0.001 0.002 0.003 0.016 1.1 · 10-4 5.7 · 10-4 7.1 · 10-4 3.0 · 10-4 0.004 0.008

Bhabha 1.7 · 10-4 0.007 0.011 0.103 1.4 · 10-7 1.4 · 10-6 2.5 · 10-6 2.9 · 10-5 2.7 · 10-5 0.001 0.014 5.0 · 10-9 7.5 · 10-8 4.1 · 10-6

di-muon 3.8 · 10-4 0.013 0.004 0.105 4.9 · 10-5 0.002 0.001 0.033 4.6 · 10-4 0.018 0.128 6.9 · 10-8 1.9 · 10-6 0.007

1.5 · 10-5 6.8 · 10-5 7.1 · 10-5 0.002 2.7 · 10-4 7.4 · 10-4 0.001 0.033 2.2 · 10-6 1.0 · 10-4 9.8 · 10-4 1.3 · 10-5 0.001 0.018
4

DATA 9.8 · 10-4 0.004 0.002 0.096 8.0 · 10-4 0.003 0.002 0.038 2.1 · 10-5 5.7 · 10-4 0.004 6.5 · 10-5 0.003 0.021

hh O.Igonkina


Particle identification and (M ,E )

The effect of PID on (M , E ) distribution can not be ignored
PID efficiency

1. One needs to divide background samples into subsamples: e.g. has modes with 0, 1, 2 real Kaons in final state; ¯ cc D + D- , D- K + - e- e or ¯ D- K + - -. 2. In addition, the correlation between PID efficiency and E is significant while subdivision on subsample is not obvious (like for uds)

0.0015

0.0005

PID efficiency

0.0025

0.002

0.001

0

-0.6

-0.4

-0.2

0

0.2

0.4 E, GeV

E after preselection
hh O.Igonkina 5


Selection

· mass of one prong side m1pr : 0.6GeV < m1pr < 1.9 GeV where m1pr = pµ k1pr + pµ 1pr + pµ , mmiss = 0 neut miss tr · 0.2 < · Total p
miss < 2.4 rad. cms > 0.2 GeV T

· no photons (E > 100MeV) on 3 prong side · no more than 3/0 photons (E > 100MeV) on 1 prong side for channels with electron/muon final state. - µ+ K - K - channel has no more than 3 unassociated photons.

Other cuts which have been considered are cut on pcms , on 13 (angle 1pr between 1 and 3-prong sides), Nbadtrk , veto on N on 3 prong side. These cuts were not found to be effective.

hh O.Igonkina

6


Selection variables
- e- K + K
EKKr

-
MPPw

- µ+ -

miss
220 200 180

-

m

EKKr

1pr

m

MPPw

6 5

1pr
100 80


60

miss

10

160 140 120

4 1 3 10
-1

100 80 40

2

60 40 20 20

1 10
-2

0

0.5

1

1.5

2

2.5

3 3.5 4 4.5 5 2 mass of 1 prong (GeV/c )

0 0

0.5

1

1.5

2 2.5 3 of missing momentum (rad)

0 0

0.5

1

1.5

2

2.5

3 3.5 4 4.5 5 mass of 1 prong (GeV/c 2 )

0 0

0.5

1

1.5

2 2.5 3 of missing momentum (rad)

EKKr

EKKr

MPPw

MPPw

7 6 5

p

cms 1pr
10

p
1

cms T

80 70 60 50

p

cms 1pr

180 160 140 120 100

p

cms T

4 3 2 1 0 0

40 80 30 60 40 20 1 2 3 4 5 6 momentum of 1 prong in CMS (GeV/c) 0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 transverse total momentum in CMS (GeV/c)

10

-1

20 10

1

2

3 4 5 6 momentum of 1 prong in CMS (GeV/c)

0

0.5

1

1.5 2 2.5 3 3.5 4 4.5 5 transverse total momentum in CMS (GeV/c)

0 0

EKKr

N

1pr

EKKr

N

25

70 60

3pr

MPPw
2200 2000 1800 1600

N

1pr

MPPw
2200 2000 1800 1600 1400 1200 1000 800 600 400 200

N

3pr

20 50 15 40 30 20 5 10 0 0

1400 1200 1000

10

800 600 400 200 0 1 2 3 4 5 6 7 8 9 N on 3 prong side 0 0 1 2 3 4 5 6 7 8 9 N on 1 prong side

0

1

2

3

4

5

6

7 8 9 N on 1 prong side

0

0

1

2

3

4

5

6

7 8 9 N on 3 prong side

hh O.Igonkina

points - data, red - signal, shade - background, lumi scaled

7


Selection efficiency



- - - - - - - - - - - - - -



e- K + K - e- K + - e- + K - e- + - µ- K + K - µ- K + - µ- + K - µ- + - e+ K - K - e+ K - - e+ - - µ+ K - K - µ+ K - - µ+ - -

Signal 0.4859 0.3672 0.3692 0.3580 0.2999 0.3015 0.2967 0.2829 0.4908 0.3707 0.3587 0.5349 0.2962 0.2895

0 0 0 0 0 0 0 0 0 0 0 0 0 0

bb .0355 .0279 .0317 .0218 .0035 .0059 .0060 .0015 .0341 .0300 .0228 .0341 .0060 .0015

0 0 0 0 0 0 0 0 0 0 0 0 0 0

. . . . . . . . . . . . . .

cc 0396 0270 0250 0062 0024 0022 0019 0006 0380 0250 0061 0380 0019 0006

uds 0.0900 0.0756 0.0740 0.0500 0.0048 0.0044 0.0043 0.0029 0.0879 0.0739 0.0501 0.0879 0.0043 0.0029

Bhabha 0.0690 0 0 0 0 0.0123 0.0145 0 0.0667 0 0 0.0667 0.0141 0

di-muon 0.0277 0.0412 0.0400 0.0526 0.0106 0.0168 0.0173 0.0214 0.0509 0.0430 0.0471 0.0509 0.0152 0.0223

0.3320 0.2008 0.1995 0.1891 0.1209 0.1241 0.1229 0.1144 0.3321 0.1991 0.1891 0.3321 0.1233 0.1148

DATA 0.0481 0.0315 0.0116 0.0058 0.0510 0.0367 0.0525 0.0926 0.0857 0.0373 0.0502 0.1667 0.0537 0.0919

hh O.Igonkina

8


M plot before fit (MC - lumi scaled, data blinded)
EKKr EPKr EPPr 22 6 3 20 18 5 2.5 16 14 12 3 1.5 10 8 2 1 6 4 2 0 -0.6 -0.4 -0.2 0 0.2 0.4 2 M (GeV/c ) 0 -0.6 -0.4 -0.2 0 0.2 0.4 2 M (GeV/c ) 0 -0.6 -0.4 -0.2 0 0.2 0.4 2 M (GeV/c ) 0 -0.6 -0.4 -0.2 0 0.2 0.4 2 M (GeV/c ) 2 3 5 6 MKKr

4

2

4

1

0.5

1

MPKr 12

MPPr

EKKw

EKPw 2.2

140 10 120

1

2 1.8

0.8 8 100 80 60 4 40 2 0.2 20 0 0 0.6

1.6 1.4 1.2 1

6

0.4

0.8 0.6 0.4 0.2

0

-0.6

-0.4

-0.2

0

0.2 0.4 2 M (GeV/c )

-0.6

-0.4

-0.2

0

0.2 0.4 2 M (GeV/c )

-0.6

-0.4

-0.2

0

0.2 0.4 2 M (GeV/c )

0

-0.6

-0.4

-0.2

0

0.2 0.4 2 M (GeV/c )

EPPw 14 12 10 8 6 4 2

MKKw 2.2 2 1.8 1.6 1.4 1.2

MKPw

MPPw 80

9 70 8 7 6 5 60 50 40 30 20 2 1 -0.6 -0.4 -0.2 0 0.2 0.4 2 M (GeV/c ) 0 -0.6 -0.4 -0.2 0 0.2 0.4 2 M (GeV/c ) 10 0

1 0.8 0.6 0.4 0.2 4 3

0

-0.6

-0.4

-0.2

0

0.2 0.4 2 M (GeV/c )

0

-0.6

-0.4

-0.2

0

0.2 0.4 2 M (GeV/c )

hh O.Igonkina

9


data fit
- e- K + K
-

- µ+ -

-

hh O.Igonkina

10


Expected sensitivity for 220 fb
Channel - e- K + K - - e- K + - - e- + K - - e- + - - µ- K + K - - µ- K + - - µ- + K - - µ- + - - e+ K - - - e+ - - - µ+ K - K - - µ+ K - - - µ+ - - Efficiency exp Bkgr Sensitivity 6.43% 2.3 1.66 · 10-7 6.36% 3.11 1.84 · 10-7 6% 0.912 1.39 · 10-7 7.3% 6.99 2.05 · 10-7 2.92% 0.222 2.48 · 10-7 3.85% 0.791 2.16 · 10-7 3.39% 0.86 2.45 · 10-7 4.48% 2.07 2.37 · 10-7 6.07% 0.709 1.37 · 10-7 7.45% 1.83 1.34 · 10-7 4.88% 0.495 1.48 · 10-7 3.42% 1.35 2.68 · 10-7 4.42% 1.1 2.07 · 10-7

-1

CLEO 6.0 · 10-6 3.8 · 10-6 6.4 · 10-6 2.2 · 10-6 15 · 10-6 7.4 · 10-6 7.5 · 10-6 8.2 · 10-6 2.1 · 10-6 1.9 · 10-6 6.0 · 10-6 7.0 · 10-6 3.4 · 10-6

Even without fine cut tuning and (M ,E ) optimization the expected sensitivity is very good.
hh O.Igonkina 11


Conclusions

· Main ingredients are in place : efficiency, background analysis, fit. · Cut optimization needs a final tuning, hopefully will go very fast · Systematic uncertainties will be adopted from analysis with correction for CM2 difference and PID-weighting · Will probably make a preliminary result for 14 channels to show at TAU2004 · The final publication (PRD) should follow very soon after, with added. h0 , h0 h+ h- and update of

hh O.Igonkina

12