The H $\rightarrow$ ZZ $\rightarrow$ l ⁺l ^-$\nu$$\bar{\nu}$ channel

 The H $\rightarrow$ ZZ decay channels where one of the vector bosons are required to decay to electrons or muons has the potential for a clean signal using the requirement of a pair of leptons with an invariant mass close to m_Z . The disadvantage is the branching ratio which for the H $\rightarrow$ ZZ $\rightarrow$ l ⁺l ^-$\nu$$\bar{\nu}$ channel is 18 times lower than the branching ratio for the H $\rightarrow$ W ⁺W ^- $\rightarrow$ l$\nu$jj decay. The decay with two neutrinos on one side of the event creates a large missing transverse energy. In fig. 8.10 the missing transverse energy is shown for the H $\rightarrow$ ZZ $\rightarrow$ l ⁺l ^-$\nu$$\bar{\nu}$ signal and the dominant t$\bar{t}$ and jZ backgrounds after the requirement of two isolated leptons with transverse momentum above 100 GeV.

  

Figure: The missing transverse momentum in H $\rightarrow$ ZZ $\rightarrow$ l ⁺l ^-$\nu$$\bar{\nu}$ events (a) and the dominant t$\bar{t}$ and jZ backgrounds in (b) after a requirement on two isolated leptons with p_T > 100 GeV.

In table 8.9 the effect of all identification cuts are shown. An efficiency for the identification of already isolated leptons of 90% has been assumed. The constraint of the lepton pair mass to m_Z has a large effect caused by the good di-lepton mass resolution of the ATLAS detector and the fact that most lepton pairs in the t$\bar{t}$ and jW events are fake pairs with large separation. With an efficiency of 80% for the signal it reduces the jZ background by a factor six and eliminates the t$\bar{t}$ and jW backgrounds.

 

$$\bar{t}$$ Cut type H VV continuum

(GeV) # / 100 fb ^{- 1} (%) # / 100 fb ^{- 1} (%) # / 100 fb ^{- 1} (%)

$$\pm$$ - - - - - -

$$\pm \pm \pm$$ p_T^lep1 > 100 78 - -

$$\pm \pm \pm$$ p_T^lep2 > 40 75 16 11

$$\pm \pm \pm$$ E_T^miss > 200 93 26 13

$$\pm \pm$$ |m_ll - m_Z| < 5 80 15 < 1.5 -

$$\pm \pm$$ E_T^tag > 30 70 14 < 1.5 -

$$\pm$$ 2 x E_T^tag > 30 29 < 1.2 - < 1.5 -

Cut type jZ jW

(GeV) # / 100 fb ^{- 1} (%) # / 100 fb ^{- 1} (%)

$$\pm \pm$$ p_T^lep1 > 100 - -

$$\pm \pm$$ p_T^lep2 > 40 59 0.6

$$\pm \pm$$ E_T^miss > 200 0.2 21

$$\pm$$ |m_ll - m_Z| < 5 13 < 2.1 -

$$\pm$$ E_T^tag > 30 23 < 2.1 -

$$\pm$$ 2 x E_T^tag > 30 53 < 2.1 -

 

The total number of signal events per year is well below the number of events in the H $\rightarrow$ W ⁺W ^- $\rightarrow$ l$\nu$jj channel but the background is almost absent. The rejection obtained with the requirement of two tag jets is limited by the effects of pile-up.

Replacing the requirement on tag jets with a track veto in the central region gives 12 signal events in a year with a background of 23 events totally dominated by the vector boson continuum. The significance is slightly worse with these cuts but will be valuable as a cross check on the jZ background.