[Paper Review] Features in the Standard Model diphoton background

Kyrylo Bondarenko, Alexey Boyarsky+2 more|arXiv (Cornell University)|Jun 30, 2016

Particle physics theoretical and experimental studies3 citations

TL;DR

This paper proposes that bump-like features in the Standard Model diphoton background—such as the 750 GeV excess observed by ATLAS and CMS—may arise from misidentified $ackslash{}$eta$ mesons decaying via $\eta \to 3\pi^0 \to 6\gamma$, whose photons are clustered into single-pixel deposits due to calorimeter granularity. Simulations show that this effect can produce a peak near 750 GeV in the diphoton invariant mass spectrum, challenging the assumption of smooth SM background.

ABSTRACT

We argue that electromagnetic decays of energetic unflavoured neutral mesons, notably $η$, mis-identified as single photons due to granularity of the electromagnetic calorimeter might create bump-like features in the diphoton invariant mass spectrum at different energies, including 750 GeV. We discuss what kind of additional analysis can exclude or confirm this hypothesis.

Motivation & Objective

• To investigate whether unflavored neutral mesons, particularly $\eta$ mesons, could produce bump-like features in the diphoton invariant mass spectrum due to detector granularity.
• To assess the hypothesis that $\eta \to 3\pi^0 \to 6\gamma$ decays are misidentified as single photons, leading to artificial peaks in the diphoton spectrum.
• To evaluate whether such effects could explain the 750 GeV diphoton excess observed by ATLAS and CMS, without invoking new physics.
• To identify experimental checks that could confirm or rule out this background origin of the observed feature.
• To highlight the need for full detector simulation and Monte Carlo analysis to quantify the effect's normalization and shape.

Proposed method

• Simulating the angular distribution of photons from $\eta \to 3\pi^0 \to 6\gamma$ decays in the $\eta$ rest frame, assuming isotropic $3\pi^0$ decay and using TGenPhaseSpace for n-body decays.
• Modeling the energy and angular distribution of $\eta$ mesons in the lab frame using a relativistic phase-space distribution with Lorentz factor $\gamma$.
• Applying a detector response model based on ATLAS and CMS electromagnetic calorimeter granularity, using pixel size thresholds ($\Delta\eta \approx 0.003-0.025$, $\Delta\phi \approx 0.1$) to simulate photon clustering.
• Convoluting the photon energy and angular distribution with the detector's pixel response to estimate the probability of misidentifying multiple photons as a single isolated signal.
• Generating the diphoton invariant mass spectrum by combining two misidentified $\eta$-derived photons using the relativistic invariant mass formula: $m_{2\eta} = \sqrt{E_1E_2 - p_1p_2\cos\alpha}$.
• Comparing the resulting spectrum with the observed 750 GeV excess, particularly in ATLAS and CMS detector configurations, noting that CMS’s broader granularity shifts the peak to lower energies (~5.8× lower).

Experimental results

Research questions

• RQ1 Can $\eta \to 3\pi^0 \to 6\gamma$ decays, when misidentified due to calorimeter granularity, produce a bump-like feature in the diphoton invariant mass spectrum?
• RQ2 What is the expected energy scale and shape of such a bump in the ATLAS and CMS detectors, given their differing granularity?
• RQ3 How does the misidentification probability depend on the $\eta$ meson energy, Lorentz boost, and detector pixel size?
• RQ4 Would stronger photon isolation cuts suppress the observed feature, and could this serve as a test of the hypothesis?
• RQ5 Is the observed 750 GeV diphoton excess in ATLAS consistent with this background effect, or does it require new physics?

Key findings

• Simulations show that $\eta \to 3\pi^0 \to 6\gamma$ decays can produce a bump-like feature in the diphoton invariant mass spectrum centered at approximately 750 GeV in the ATLAS detector configuration.
• The peak arises from the convolution of the growing misidentification probability with the falling number of high-energy $\eta$ mesons, leading to a non-monotonic background shape.
• The position of the bump depends on the detector’s electromagnetic calorimeter granularity, with CMS’s coarser resolution shifting the peak to about 5.8 times lower energy than in ATLAS.
• The fraction of $\eta$-meson energy carried by photons deviating from the original direction is wider than in the two-photon $\pi^0$ decay, but most energy is still carried by photons aligned with the $\eta$-meson’s boost direction.
• The probability of depositing a large fraction of $\eta$-decay energy in a single pixel is significant, especially for photons near the direction of the $\eta$ meson, increasing misidentification likelihood.
• Without full detector simulation and Monte Carlo analysis, the normalization of the peak cannot be quantified, but the qualitative effect is plausible and potentially relevant to the 750 GeV excess.