AI 中文总结
研究由标量场与辐射早期相互作用推动的光子扇区现象学扩展，通过在CLASS中实现修改后的光子标度并结合相关数据进行分析，得出常数$\epsilon$扩展受晚期数据组合青睐的结论。
AI 中文摘要
我们研究了由标量场分量与辐射之间的早期相互作用所推动的光子扇区的现象学扩展。该模型由常数参数$\epsilon$描述，它衡量光子能量密度和CMB温度-红移关系与其标准绝热演化的偏离。当$\epsilon = 0$时恢复标准光子扇区。我们在CLASS中实现修改后的光子标度并验证数值背景演化与解析的常数$\epsilon$预测一致。接着研究了复合历史、可见度函数和CMB温度谱的诊断响应。这些诊断表明小的$\epsilon$值会改变复合历史并修改CMB温度各向异性谱的声学峰结构。这些CMB输出仅用作一致性和响应诊断而非完整的CMB似然约束。作为初步统计应用，我们将修改后的CLASS实现与MontePython结合，使用Pantheon+SH0ES超新星数据和BAO距离测量来约束模型。晚期分析给出$\epsilon = 0.0230 \pm 0.0065$，相对于固定$\epsilon = 0$基线，有效$\Delta\chi^2 = -10.518$，$\Delta$AIC = -8.518，改善了最佳拟合似然。这些结果表明根据AIC标准，常数$\epsilon$扩展受此初步晚期数据组合青睐。
英文摘要
We investigate the constant-$ε$ limit of a photon-sector modification derived from an early-time interaction between a scalar field and radiation. Starting from the underlying scalar--photon coupling, the parameter $ε$ quantifies the departure of the photon energy density from the standard adiabatic scaling, leading in the constant-$ε$ limit to $ρ_γ\propto a^{-4+ε}$ and to the modified CMB temperature--redshift relation $T(z)=T_{0}(1+z)^{1-ε/4}$. The standard photon sector is recovered for $ε=0$. We implement the corresponding constant-$ε$ background evolution in CLASS and verify that the numerical photon-density evolution reproduces the analytic scaling. We then examine the resulting changes in the recombination history, visibility function and CMB temperature-anisotropy spectrum. These CMB spectra are used only as diagnostic outputs since the coupled scalar--photon perturbation equations are not implemented self-consistently in the numerical calculation and a full CMB likelihood analysis is beyond the scope of the present work. As a preliminary statistical application, we interface the modified CLASS implementation with MontePython and analyze the Pantheon+SH0ES supernova likelihood together with BAO measurements. Within the adopted constant-$ε$ framework, prior choices, data combination and fixed early Universe parameter setup, we obtain $ε=0.0230\pm0.0065$. The posterior is therefore centered on a positive value within this restricted analysis. Because the BAO predictions depend on the model-dependent drag scale and several early-Universe quantities are held fixed, this result should be interpreted as a conditional and preliminary constraint rather than as a full CMB-level determination or evidence for a resolution of the Hubble tension.
Comments20 pages, 4 figs
