ALP-ine quests at the LHC: hunting axion-like particles via peaks and dips in $t \bar{t}$ production
- Anuar, Afiq et al
- arXiv:2404.19014DESY-24-059IFT-UAM/CSIC-24-042KA-TP-06-2024MITP-24-044
| The background-subtracted inclusive $pp \to \ttbar$ cross section, including resonant $pp \to a$ production and ALP/SM interference, $\sigma_{t \bar t}^{\rm tot} - \sigma_{t \bar t}^{\rm SM}$, is shown in the plane of the ALP couplings $\cG/f_a$ and $\ct/f_a$ for different ALP masses and relative widths of $400 \gev$ and 2.5\% (left) as well as $800 \gev$ and 5.0\% (right). Negative values of $\ct$ are not shown as the cross section is symmetric under the sign change of both $\ct$ and $\cG$, see Eq.~\eqref{eq:xSec_para}. |
| The background-subtracted inclusive $pp \to \ttbar$ cross section, including resonant $pp \to a$ production and ALP/SM interference, $\sigma_{t \bar t}^{\rm tot} - \sigma_{t \bar t}^{\rm SM}$, is shown in the plane of the ALP couplings $\cG/f_a$ and $\ct/f_a$ for different ALP masses and relative widths of $400 \gev$ and 2.5\% (left) as well as $800 \gev$ and 5.0\% (right). Negative values of $\ct$ are not shown as the cross section is symmetric under the sign change of both $\ct$ and $\cG$, see Eq.~\eqref{eq:xSec_para}. |
| Differential distribution in $\mtt$ for the SM (top) and for a pseudoscalar Higgs boson with $m_A = 400 \gev$, 4.0\% width and a coupling strength of $\gAtt = 0.9$, where the SM background has been subtracted (bottom). The left plot shows the distribution inclusive in the variable $\chel$, while the right plot shows it after selecting only events with $\chel > 0.6$. Our smeared prediction (blue) is compared to the CMS simulation taken from \cite{CMS:2019pzc} (red). All predictions are shown for an integrated luminosity of $35.9 \, \fb$. The gray bands show the expected statistical uncertainty from the SM $\ttbar$ background. |
| Left: The effect of a top-quark mass variation of $\pm 1 \gev$ in the SM $\ttbar$ background (green and blue lines; the central value is indicated by the black line in the upper plot, all displayed SM curves are normalized such that they yield the same total cross section) compared to the effect of a pseudoscalar Higgs boson with $m_A = 400 \gev$, 4\% width and $\gAtt = 0.9$ (red dashed line). The gray band in the lower plot, where the SM background has been subtracted, indicates the statistical uncertainty for an integrated luminosity of $35.9 \, \fb$. Right: The expected significance for the pseudoscalar Higgs boson as a function of its coupling $\gAtt$ for the full set of systematic uncertainties, the full set except for the top-quark mass uncertainty, and for the case where only statistical uncertainties are taken into account. The expected significance reported by CMS for $\gAtt = 0.9$ is shown as the red star. |
| Left: The effect of a top-quark mass variation of $\pm 1 \gev$ in the SM $\ttbar$ background (green and blue lines; the central value is indicated by the black line in the upper plot, all displayed SM curves are normalized such that they yield the same total cross section) compared to the effect of a pseudoscalar Higgs boson with $m_A = 400 \gev$, 4\% width and $\gAtt = 0.9$ (red dashed line). The gray band in the lower plot, where the SM background has been subtracted, indicates the statistical uncertainty for an integrated luminosity of $35.9 \, \fb$. Right: The expected significance for the pseudoscalar Higgs boson as a function of its coupling $\gAtt$ for the full set of systematic uncertainties, the full set except for the top-quark mass uncertainty, and for the case where only statistical uncertainties are taken into account. The expected significance reported by CMS for $\gAtt = 0.9$ is shown as the red star. |
| Limit on the coupling of an ALP to the top quark, $\ct/f_a$, in the case $\cG = 0$, translated from \citere{CMS:2019pzc}, for a relative ALP width of 2.5\% (left) and 5\% (right). The hatched band shows the region in which the sum of the predicted partial widths for the $a \to \ttbar, gg,\gamma\gamma$ decays exceeds the assumed total width. |
| Limit on the coupling of an ALP to the top quark, $\ct/f_a$, in the case $\cG = 0$, translated from \citere{CMS:2019pzc}, for a relative ALP width of 2.5\% (left) and 5\% (right). The hatched band shows the region in which the sum of the predicted partial widths for the $a \to \ttbar, gg,\gamma\gamma$ decays exceeds the assumed total width. |
| Differential distribution in $\mtt$ for an ALP with different values of $\cG$ and $\ct$ and a pseudoscalar Higgs boson with different values of $\gAtt$, both with a mass of $400\gev$ and a total width of 2.5\%. The couplings $\cG$, $\ct$ and $\gAtt$ are chosen in the considered benchmark scenarios such that the ALP and the pseudoscalar Higgs boson have the same integrated cross section in a given panel. Event counts are shown for integrated luminosities corresponding to Run 2 ($138 \, \fb$, left axis) and the HL-LHC ($3 \, \ab$, right axis). The gray bands indicate the expected statistical uncertainties on the SM background for the two integrated luminosities. |
| Projected expected limits on the ALP couplings $\cG/f_a$ and $\ct/f_a$ as obtained by the maximum likelihood fit for three different integrated luminosities, corresponding to LHC Run 2 (red), Run 2+3 (blue) and HL-LHC (green), and for $m_a = 400 \gev$, $\Gamma_a/m_a = 2.5\%$ (left) and $800 \gev$, $\Gamma_a/m_a = 5.0\%$ (right). The shaded bands show the variations of the expected limit by one standard deviation. For all cases, all systematic uncertainties are included (scaled by a factor of 0.5 for the HL-LHC) except for the top-quark mass uncertainty. The region in which the assumed ALP width is lower than the total predicted width, taking into account the $a \to \ttbar$, $a \to gg$ and $a \to \gamma\gamma$ decays, is shown as the gray shaded area. In the left plot, the benchmark points from \reffi{fig:distributions}, with the same color coding, are shown as stars. |
| Projected expected limits on the ALP couplings $\cG/f_a$ and $\ct/f_a$ as obtained by the maximum likelihood fit for three different integrated luminosities, corresponding to LHC Run 2 (red), Run 2+3 (blue) and HL-LHC (green), and for $m_a = 400 \gev$, $\Gamma_a/m_a = 2.5\%$ (left) and $800 \gev$, $\Gamma_a/m_a = 5.0\%$ (right). The shaded bands show the variations of the expected limit by one standard deviation. For all cases, all systematic uncertainties are included (scaled by a factor of 0.5 for the HL-LHC) except for the top-quark mass uncertainty. The region in which the assumed ALP width is lower than the total predicted width, taking into account the $a \to \ttbar$, $a \to gg$ and $a \to \gamma\gamma$ decays, is shown as the gray shaded area. In the left plot, the benchmark points from \reffi{fig:distributions}, with the same color coding, are shown as stars. |
| Projected expected limits on the ALP couplings $\cG/f_a$ and $\ct/f_a$ that are obtained as in \reffi{fig:projectedlimits}, for $m_a = 400 \gev$, $\Gamma_a/m_a = 2.5\%$ (left plot) and $\ma = 800 \gev$, $\Gamma_a/m_a = 5\%$ (right plot) and for three different treatments of the uncertainties as defined in \refse{sec:syst_uncert}: with all systematic uncertainties including the top-quark mass uncertainty (blue), excluding the top-quark mass uncertainty but keeping other uncertainties (red), and considering statistical uncertainties only (gray). All limits are shown for an integrated luminosity of $138 \, \fb$, corresponding to Run 2. |
| Projected expected limits on the ALP couplings $\cG/f_a$ and $\ct/f_a$ that are obtained as in \reffi{fig:projectedlimits}, for $m_a = 400 \gev$, $\Gamma_a/m_a = 2.5\%$ (left plot) and $\ma = 800 \gev$, $\Gamma_a/m_a = 5\%$ (right plot) and for three different treatments of the uncertainties as defined in \refse{sec:syst_uncert}: with all systematic uncertainties including the top-quark mass uncertainty (blue), excluding the top-quark mass uncertainty but keeping other uncertainties (red), and considering statistical uncertainties only (gray). All limits are shown for an integrated luminosity of $138 \, \fb$, corresponding to Run 2. |
| Limits on the ALP Wilson coefficients $\ct/f_a$ and $\cG/f_a$ from various sources: The red line shows our result for the projected expected 95\% CL upper limit on the ALP couplings $\cG/f_a$ and $\ct/f_a$ from ALP searches in the invariant $t \bar t$ mass distribution and angular correlations for an ALP with a mass of $400\gev$ and relative width of $2.5\%$ for the case where all statistical and systematic uncertainties are included except for the top-quark mass uncertainty (see Fig.~\protect\ref{fig:projectedlimits}). The red dashed lines indicate the $\pm 1\sigma$ uncertainty band. The green shaded areas are excluded at 95\% CL from LHC searches for narrow di-photon resonances performed by ATLAS~\cite{ATLAS:2021uiz}, and the blue shaded area is excluded from the CMS cross section measurement for the production of four top quarks~\cite{CMS:2023ftu}, where both analyses are based on the full Run~2 dataset. Values of $c_t / f_a$ above the gray hatched line are excluded by ALP--SMEFT interference effects~\cite{Biekotter:2023mpd} (see text for details). Coupling values for which the predicted sum of the partial widths for $a \to \ttbar, gg, \gamma\gamma$ is larger than the assumed total width of the ALP are indicated with the gray shaded area. The black dot-dashed line represents the couplings corresponding to a top-philic ALP. |