[论文解读] Charged-lepton-flavour violation at the LHC: a search for $Z o e au/\mu au$ decays with the ATLAS detector
本论文利用大型强子对撞机(LHC)ATLAS探测器在√s = 13 TeV下收集的139 fb⁻¹质子-质子碰撞数据,搜索了Z玻色子衰变为底夸克的τ轻子与更轻的带电轻子(e或μ)的轻子味违反衰变。通过采用新型神经网络分类器和分箱最大似然拟合方法,该研究设定了迄今为止最严格的限制,观测到的分支分数上限分别为8.1 × 10⁻⁶(eτ)和9.5 × 10⁻⁶(μτ),置信水平为95%,超越了LEP实验先前的约束。
In the Standard Model of particle physics, leptons are key building blocks of matter and come in three families (flavours). Leptons of different flavours have the same properties, except for their mass. In addition, the number of leptons in each family is conserved in interactions. Such conservation is known as lepton flavour conservation, and no fundamental principles impose it. Since the formulation of the Standard Model, the observation of flavour oscillations among neutrinos (the neutral leptons) has demonstrated that neutrinos have mass and in neutrino weak interactions the lepton flavour is not conserved. To date, there is no experimental evidence that lepton flavour violation occurs in interactions between charged leptons, and an observation of such a phenomenon would be an exciting sign of new particles or new type of interactions beyond the Standard Model. The ATLAS experiment at the Large Hadron Collider at CERN sets a new constraint on lepton-flavour-violating effects in weak interactions, searching for $Z$-boson decays into a $ au$-lepton and another lepton of different flavour ($e$ or $\mu$) with opposite electric charge. The branching fractions for these decays are now measured by the ATLAS experiment to be less than $8.1 imes10^{-6}$ ($e au$) and $9.5 imes10^{-6}$ ($\mu au$) at 95% confidence level, using 139 fb$^{-1}$ of proton-proton collision data at centre-of-mass energy $\sqrt{s}=13$ TeV and 20.3 fb$^{-1}$ at $\sqrt{s}=8$ TeV. These results supersede the best limits set by the LEP experiments more than two decades ago.
研究动机与目标
- 搜索Z玻色子衰变为τ轻子与更轻的带电轻子(e或μ)的带电轻子味违反衰变,此类过程若被观测到将表明标准模型之外的新物理。
- 通过采用先进的机器学习技术抑制背景,提高对稀有Z玻色子衰变的探测灵敏度。
- 设定迄今为止最严格的Z → eτ和Z → μτ分支分数实验上限,超越LEP实验先前的约束。
- 检验通过重中微子或其他新物理机制预测轻子味违反的理论模型。
提出的方法
- 本分析使用了LHC Run 2期间ATLAS探测器在√s = 13 TeV下收集的139 fb⁻¹质子-质子碰撞数据。
- 事件选择包含一个强子衰变的τ轻子(τhad-vis),通过使用电磁量能器和轨迹信息的循环神经网络进行识别。
- 电子和μ子候选事例通过中等识别标准和严格隔离标准选出,要求pT > 30 GeV且|η| < 2.5。
- 训练神经网络分类器以最优方式区分信号事例与主要背景(包括Z → ℓℓ和Z → ττ衰变)。
- 采用分箱最大似然拟合方法提取信号事例数,并在95%置信水平下设定分支分数上限。
- μτ通道的结果与先前LHC Run 1分析(√s = 8 TeV,20.3 fb⁻¹)结合,以提升探测灵敏度。
实验结果
研究问题
- RQ1Z → eτ衰变的分支分数上限是多少?与以往实验相比如何?
- RQ2Z → μτ衰变的分支分数上限是多少?是否优于早期LEP实验的约束?
- RQ3基于神经网络的分类器是否能显著提升对稀有Z玻色子衰变的探测灵敏度,相比传统方法?
- RQ4将Run 1与Run 2数据结合,能在多大程度上提升对轻子味违反Z衰变的探测灵敏度?
主要发现
- 观测到的Z → eτ衰变分支分数上限为8.1 × 10⁻⁶(95%置信水平)。
- 观测到的Z → μτ衰变分支分数上限为9.5 × 10⁻⁶(95%置信水平)。
- 这些上限为迄今最严格,超越了先前LEP实验的约束(eτ为9.8 × 10⁻⁶,μτ为1.2 × 10⁻⁵)。
- 采用循环神经网络进行τhad-vis识别,显著提升了信号与背景的区分能力。
- Run 1与Run 2数据的结合增强了探测灵敏度,尤其在μτ通道中效果明显。
- 结果对通过重中微子或其他新物理机制预测轻子味违反的理论模型提供了强有力的约束。
更好的研究,从现在开始
从论文设计到论文写作,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。