宇宙有多“贪心”？早期黑洞增长的熵账本与因果包络
How greedy is the Universe? An entropy ledger and a causal envelope for early black hole growth
AI总结：
该研究定量检验宇宙是否以黑洞为优先熵产生通道，发现黑洞熵产生远超辐射通道但远低于因果包络，宇宙未最大化熵产生，否定了相关强最大熵产生原理。
AI中文摘要：
可观测宇宙的熵在15个数量级上由超大质量黑洞的视界主导，引力坍缩形成黑洞常被认为是宇宙产生熵的优先通道。我们对这一观点进行定量检验，分析结果基本否定了该观点，共得到三项结果。第一，熵产生历史：按通道分解的熵产生率dS/dt(z)，来自恒星形成历史和Soltan归一化的吸积历史；第一代种子之后，黑洞视界增长超过所有辐射通道约10^16-17倍，总率在z≈1-2附近达到峰值后下降，晚峰源于结构特性，因为视界熵产生按吸积体质量加权，而最大质量吸积体形成最晚。第二，因果流体动力学包络：由因果性和气体动力学允许的最快熵产生轨迹，原子冷却晕中吸积率为M̃~c_s^3/G，在z≈8-10时达到10^(7±1)M⊙且无辐射节流，其现象学（遮蔽、辐射低效、X射线弱、红色）与JWST（詹姆斯·韦布空间望远镜）小红点高度相似；实现率与包络率的比值定义为效率F(z)，在4≤z≤12范围内始终介于10^-7和10^-4之间，即使在JWST时代最宽松的区间内也保持该范围。第三，量级级别的不可能结果：按熵增益对坍缩通道进行玻尔兹曼加权、对自引力系统应用耗散适应，以及宇宙学中任何强最大熵产生原理，在主导阶均失效，失效因子最高达10^16。宇宙的熵绝大多数由黑洞产生，但始终远低于自身的包络；无论选择何种宇宙结构，都不会最大化熵产生。
英文摘要：
The entropy of the observable Universe is dominated, by fifteen orders of magnitude, by the horizons of supermassive black holes, and gravitational collapse into black holes has often been proposed as the Universe's preferred channel of entropy production. We examine this suggestion quantitatively, and the analysis largely refutes it. Three results are presented. First, an entropy production history: the rate $dS/dt(z)$ resolved by channel, from the star formation history and a Soltan-normalized accretion history. Black hole horizon growth exceeds all radiative channels by a factor $\sim 10^{16-17}$ after the first seeds; the total rate peaked near $z\approx 1$--$2$ and has since declined; the late peak is structural, because horizon entropy production weights accretion by the accretor's mass and the most massive accretors are assembled last. Second, a causal-hydrodynamic envelope: the fastest entropy-producing trajectory permitted by causality and gas dynamics, $\dot M \sim c_s^3/G$ in atomic-cooling halos, which reaches $10^{7\pm1}\,M_\odot$ by $z\approx 8$--$10$ without radiative throttling, and whose phenomenology (obscured, radiatively inefficient, X-ray weak, red) closely resembles the JWST little red dots. The realized-to-envelope ratio defines an efficiency $F(z)$, which remains between $10^{-7}$ and $10^{-4}$ throughout $4\le z\le 12$, even under the most generous JWST-era bracket. Third, order-of-magnitude no-go results: Boltzmann weighting of collapse channels by entropy gain, dissipative adaptation applied to self-gravitating systems, and any strong maximum-entropy-production principle in cosmology all fail at leading order, by factors of up to $10^{16}$. The Universe produces entropy overwhelmingly through black holes, yet remains far below its own envelope at all times. Whatever selects cosmic structure, it does not maximize entropy production.
