Monotop signature from the supersymmetric $t\overline{t}H$ channel

We point out that a distinctive monotop signature is present in natural supersymmetry scenarios when a scalar top quark and Higgsinos are almost mass degenerate. This signature originates from a supersymmetric counterpart of the $t\overline{t}H$ process, i.e. $pp\to \tilde{t}t\tilde{h}$. Unlike monojet signatures exploiting initial state radiation, this channel can be regarded as a clear signature of a light stop and Higgsinos, allowing a direct probe of the stop and neutralino sectors. The production rate of this channel largely depends on the up-type Higgsino components in the neutralinos while the stop sector is sensitive to angular distributions of the top-quark’s decay products. We develop an optimal search strategy to capture the supersymmetric $t\overline{t}H$ process and find that a high luminosity LHC can probe the stop and Higgsino sectors with $m_{{\tilde{t}}_1}\lesssim 380\mathrm{GeV}$ and $m_{{\tilde{t}}_1}-m_{{\tilde{\chi }}_1^0}\lesssim m_W$. Additionally, we propose a kinematic variable with which one can measure the stop mixing in this channel.

Figure 1

Monojet event topologies channel from ${\tilde{t}}_1$ pair production (left) and monotop from supersymmetric $t\overline{t}H$ process (right). The grey dashed lines represent invisible particles, while the thin grey lines represent particles that are too soft to be observed. The strong coupling and the top Yukawa coupling are denoted as ${\alpha }_s$ and $Y_t$, respectively.

Figure 2

Representative Feynman diagrams for the supersymmetric $t\overline{t}H$ process. The $q\overline{q}$ initial states are also possible for the latter two diagrams. The red dots denote the stop-top-Higgsino interaction. The stop propagator in the second diagram has to be far off shell in our parameter region $m_{{\tilde{t}}_1}-m_{{\tilde{\chi }}_1^0}<m_W$; hence it is clearly separated from the stop pair production.

Figure 3

LO cross section of the supersymmetric $t\overline{t}H$ process ($pp\to {\tilde{t}}_{1}t{\tilde{\chi }}_1^0$ and ${\tilde{t}}_{1}t{\tilde{\chi }}_2^0$ are combined) in the pure Higgsino limit at the 8 (red dashed line), 13 (red solid line), and 14 TeV (red dash-dotted line) LHC. The parameters are fixed as $\mathrm{\Delta }{m}_{{\tilde{t}}_1-{\tilde{\chi }}_1^0}=10\mathrm{GeV}$, $m_{{\tilde{\chi }}_2^0}=m_{{\tilde{\chi }}_1^0}+5\mathrm{GeV}$, $\cos {\theta }_{\tilde{t}}=1$, and $\tan \beta =20$. These LO cross sections are compared with the NLO cross sections of the ${\tilde{t}}_1$ pair production (blue solid line) and the Standard Model $t\overline{t}H$ production (black solid line) at the 13 TeV LHC.

Figure 4

Normalized transverse momentum distributions for the $b$ quark from the ${\tilde{t}}_1\to b{\tilde{\chi }}_1^{\pm }$ decay (left panel) and the fermions from the subsequent ${\tilde{\chi }}_1^{\pm }\to f{\overline{f}}^{\prime }{\tilde{\chi }}_1^0$ decay (right panel) at parton level. In the left panel we fix the chargino-neutralino mass difference to $\mathrm{\Delta }{m}_{{\tilde{\chi }}_1^{\pm }-{\tilde{\chi }}_1^0}=3\mathrm{GeV}$ and vary the stop-neutralino mass difference as $\mathrm{\Delta }{m}_{{\tilde{t}}_1-{\tilde{\chi }}_1^0}=8$, 15, and 50 GeV, whereas on the right panel we fix $\mathrm{\Delta }{m}_{{\tilde{t}}_1-{\tilde{\chi }}_1^0}=45\mathrm{GeV}$ and scan $\mathrm{\Delta }{m}_{{\tilde{\chi }}_1^{\pm }-{\tilde{\chi }}_1^0}=3$, 6, and 9 GeV. We assume $m_{{\tilde{t}}_1}=317\mathrm{GeV}$ for both panels.

Figure 5

Left: Normalized $m_{bl}$ distributions for the signal ${\tilde{t}}_{1}t{\tilde{\chi }}_{1(2)}^0$ with $\mathrm{\Delta }{m}_{{\tilde{t}}_1-{\tilde{\chi }}_1^0}=8\mathrm{GeV}$ (red solid line) and 45 GeV (blue solid line), $t\overline{t}$ (black solid line), $tW$ (black dotted line), and $W+b\overline{b}$ (blue dotted line) samples after the baseline selection. Right: Transverse mass $m_T$ distributions after the baseline and $m_{b\ell }<150\mathrm{GeV}$ cuts expected at the 13 TeV LHC. The line types and colors are assigned in the same way as in the left panel apart from the $tW$, for which the black solid line is used. In this plot the contributions of the $t\overline{t}$ and $tW$ where one or two $W$ (and $t$) decay(s) leptonically (including $\tau$) are also shown, which are $t{\overline{t}}_{2l}$ (black dashed line), $t{\overline{t}}_{1l}$ (black dotted line), and $t{W}_{2l}$ (black dot-dashed line). For both plots the signal points have $m_{{\tilde{t}}_1}=317\mathrm{GeV}$ and $m_{{\tilde{\chi }}_1^{\pm }}=m_{{\tilde{\chi }}_2^0}=m_{{\tilde{\chi }}_1^0}+3\mathrm{GeV}$.

Figure 6

Missing energy distribution ${\cancel{E}}_T$ for the signal ${\tilde{t}}_{1}t{\tilde{\chi }}_{1(2)}^0$ with $\mathrm{\Delta }{m}_{{\tilde{t}}_1-{\tilde{\chi }}_1^0}=8\mathrm{GeV}$ (red solid line) and 45 GeV (blue solid line) and the total background (black solid line) after the selection cuts Eqs. (9), (10), and (11). The breakdowns of the total background are also shown: $t\overline{t}$ (blue dotted line), $tW$ (black dash-dotted line), and $tZ$ (black dotted line). The distributions are normalized to the number of expected events at the 13 TeV LHC with $3{\mathrm{ab}}^{-1}$ integrated luminosity. The signal points are assumed to have $m_{{\tilde{t}}_1}=317\mathrm{GeV}$ and $m_{{\tilde{\chi }}_1^{\pm }}=m_{{\tilde{\chi }}_2^0}=m_{{\tilde{\chi }}_1^0}+3\mathrm{GeV}$. The last bin is an overflow bin.

Figure 7

The LO cross section in the $(m_{{\tilde{t}}_1},m_{{\tilde{\chi }}_1^0})$ plane for the ${\tilde{t}}_1={\tilde{t}}_L$ (left) and ${\tilde{t}}_1={\tilde{t}}_R$ (right) cases.

Figure 8

The signal efficiency of SR2 in the $(m_{{\tilde{t}}_1},m_{{\tilde{\chi }}_1^0})$ plane for the ${\tilde{t}}_1={\tilde{t}}_L$ (left) and ${\tilde{t}}_1={\tilde{t}}_R$ (right) cases.

Figure 9

The $2\text{−}\sigma$ sensitivities expected at the 13 TeV high luminosity LHC with $\int \mathcal{L}dt=3{\mathrm{ab}}^{-1}$ for $\mathcal{R}=0.5$ (dark-pink region), 0.75 (medium-pink region), and 1 (light-pink region) for the ${\tilde{t}}_L$ (top) and ${\tilde{t}}_R$ (bottom) cases. In deriving these sensitivities, only the signal regions with more than three expected signal events and $S/B>0.1$ are considered at each parameter point and $\mathcal{R}$. The signal region with the largest $S/\sqrt{B}$ is then used to derive the sensitivity. The current 95% C.L. excluded region is filled by grey. The region surrounded by the blue curve is obtained from the 13 TeV ATLAS $\mathrm{di}\text{−}b\text{−}\mathrm{jet}$ analysis with $\int \mathcal{L}dt=3.2{\mathrm{fb}}^{-1}$ [75]. The regions with dark and light green boundaries are excluded by the ATLAS [77] and CMS [74] monojet searches with Run 1 data corresponding to $\int \mathcal{L}dt\simeq 20{\mathrm{fb}}^{-1}$.

Figure 10

The distribution of the $p_T$ asymmetry, $\mathcal{A}$, at $(m_{{\tilde{t}}_1},m_{{\tilde{\chi }}_1^0})=(317,309)\mathrm{GeV}$. The blue and red histograms correspond to the ${\tilde{t}}_1={\tilde{t}}_L$ and ${\tilde{t}}_1={\tilde{t}}_R$, respectively. The events satisfy the selection cuts described in Eqs. (9), (10), and (11). The events in the left and right plots additionally satisfy $E_T^{\mathrm{miss}}/\mathrm{GeV}>100$ and 400, respectively.

Figure 11

The most sensitive signal region (with the largest $S/\sqrt{B}$) for each parameter point and for $\mathcal{R}=0.5$ (left panel), 0.75 (center panel), and 1 (right panel). The top and bottom panels correspond to the ${\tilde{t}}_1={\tilde{t}}_L$ and ${\tilde{t}}_1={\tilde{t}}_R$ cases, respectively. The empty circles represent the parameter point where none of the signal regions satisfies the consistence criteria that the signal contribution must be greater than three and $S/B>0.1$.