Skip to content
#small language model Open access

The Energy Fluctuation of Light Is a Particle Term Plus a Wave Term, and the Two Are Equal Exactly When the Occupation Number Is 1 ── in the energy fluctuation of a single mode, the particle term divided by the wave term equals the reciprocal of the occupation number and becomes 1 at hν = kT ln 2; at the Planck peak the particle term is 15.8 times the wave term, and for visible sunlight 91.5 times ── the separator is whether the occupation number is above or below 1 ── [Paper 1022]

Sep 2026 · Zenodo (CERN European Organization for Nuclear Research)

Abstract

In 1909 Einstein showed that the energy fluctuation of a single mode of black-body radiation is the sum of a term like that of light quanta and a term like that of classical waves. This paper checks by three routes, the formula, the distribution of the occupation number and thermodynamics, that the ratio of the two terms is fixed by one number, the reciprocal of the occupation number, and counts where the two are equal and on which side the Planck peak and sunlight lie. No new theorem or law is claimed, and no interpretation of wave-particle duality is offered. Scope of this paper (scope note): No new theorem or law is claimed──that the energy fluctuation of black-body radiation is the sum of two terms (Einstein 1909), Planck’s radiation law (Planck 1901), the light-quantum hypothesis (Einstein 1905), that the occupation number of radiation is 1/(e to the power x minus 1), that Wien’s law is the limit of small occupation and the Rayleigh-Jeans law that of large occupation, and that the fluctuation can be written kT² dU/dT are all known. Duality is not interpreted──it is not said that light is "both" or "either"; what is said is only the computation that the ratio of the two terms is the reciprocal of the occupation number. The identity of photons is not derived──that the occupation number follows a geometric distribution is taken as given; Bose statistics itself is not derived. Photon correlations are not treated──correlations between two detectors (Hanbury Brown and Twiss), bunching, and the statistics of real sources (the Poisson distribution of a laser, thermal sources) are not the subject; only one mode in thermal equilibrium is seen. The values are examples──5778 K, 2.725 K, 300 K and 6000 K, and 550 nm, 10 μm, 1 μm and 160.2 GHz are representative values for comparison; no precision for real sources is claimed. Relation to earlier papers: Paper 1005 showed that removing stimulated emission from Einstein’s ledger makes the balancing radiation follow Wien’s law──here it is counted that the fluctuation of that Wien’s law has only the particle term, the wave term vanishing on the same side as stimulated emission. Paper 41 showed that the exponent forced by flux and the phase that sets the fringes are different roots──here, within one formula, which of the particle term and the wave term wins is counted by the occupation number alone. Paper 1011 showed that the wait for the first photoelectron is written by a Poisson process of photon arrivals──here it is counted that the occupation number of visible light is 0.010926, so that those arrivals lie on the particle side. Paper 902 showed that the 4π in the coefficient of black-body radiation is the area of the unit sphere in k space──here the number of modes is left alone and only the fluctuation within one mode is seen. What is added is checking by three routes that the ratio of the two terms is the reciprocal of the occupation number, obtaining by bisection that the crossing is at ln 2 with occupation number 1, lining up the Planck peak, visible sunlight and the cosmic background by this ratio, and placing Wien and Rayleigh-Jeans in numbers as one-sided terms. First, and this is the core. The particle term divided by the wave term is the reciprocal of the occupation number, and the two are equal exactly when the occupation number is 1──with u the mean energy of one mode and hν the energy of a light quantum, the mean square fluctuation is hν·u + u², and the first term over the second, hν/u, equals e to the power x minus 1 (x = hν/kT). Bisection puts the x at which this is 1 at 0.693147 = ln 2, where the occupation number is 1.000000 and the particle share is 0.500000 (Sections 2 and 3). Second, three routes return the same numbers──at x = 0.5, ln 2, 2.821439 and 5, Einstein’s formula, the variance of the geometric distribution of the occupation number summed directly, and the thermodynamic kT² dU/dT by numerical differentiation all gave 3.917698089, 2.000000000, 0.067292322 and 0.006829673 (Section 2). Third, the Planck peak lies on the particle side──at the peak per unit frequency, x = 2.821439, the particle term is 15.801016 times the wave term; at the peak per unit wavelength, x = 4.965114, it is 142.324918 times; at the 160.2 GHz peak of the cosmic background it is 15.800801 times (Section 4). Fourth, visible sunlight lies further on the particle side, and equality comes in the infrared──at 550 nm and 5778 K, x = 4.527445, the ratio is 91.521904 and the occupation number 0.010926; the two terms are equal at 3.592448 μm. For the 2.725 K background the equality is at 7.617307 mm (39.356749 GHz), and for a 300 K body at 69.19 μm (Section 4). Fifth, Wien’s law carries only the particle term and the Rayleigh-Jeans law only the wave term──at x = 8 the fluctuation from Wien’s law differs from the Planck sum by a relative -6.708x10^-4, and at x = 0.05 the fluctuation from the Rayleigh-Jeans law differs by 0.000208; each approaches the sum with one term alone (Section 5). Sixth, the separator is whether the occupation number is above or below 1──if a mode holds on average more than one light quantum the wave term wins, if fewer than one the particle term wins. Not whether light is a wave or a particle, but how full the mode is, decides the shape of the fluctuation (Section 6). The energy fluctuation of light is a particle term plus a wave term, and the two are equal exactly when the occupation number is 1. In the fluctuation hν·u + u² of one mode, the particle term over the wave term equals the reciprocal of the occupation number and becomes 1 at hν = kT ln 2. At the Planck peak the particle term is 15.801016 times the wave term, for visible sunlight 91.521904 times, and equality lies at 3.592448 μm for the Sun and 7.617307 mm for the background──Wien’s law carries only the particle term and the Rayleigh-Jeans law only the wave term. The separator is whether the occupation number is above or below 1. Placed among the earlier papers──Paper 1005 showed that removing stimulated emission gives Wien’s law, Paper 41 that the exponent of flux and the phase of the fringes are different roots, and Paper 1011 that photoelectron arrivals are written by a Poisson process. Here it was counted that which of the two terms of one formula wins is fixed by the occupation number alone. To be honest──nothing has been said about what light is. On the making of this work: The ideas and content of this work stem from the author's own considerations. Assistance from an AI (a large language model) was used for structuring, English translation, and checking the algebra. Any remaining errors or misinterpretations are solely the author's. Feedback and corrections are sincerely appreciated. Keywords: energy fluctuation of radiation, light quantum, Planck's law, Wien's law, Rayleigh-Jeans law, occupation number. ----- アインシュタインは 1909 年に、黒体放射の一つのモードのエネルギーのゆらぎが、光量子のような項と古典的な波のような項の和になることを示した。本稿は、その二つの項の比が占有数の逆数という一つの数で決まることを、式・占有数の分布・熱力学の三つの道で確かめ、二つが等しくなる点と、プランクの山や太陽の光がどちらの側にあるかを数える。新しい定理も法則も主張せず、波と粒子の二重性の解釈もしない。 本稿の射程(射程注記):新しい定理も法則も主張しない──黒体放射のエネルギーのゆらぎが二つの項の和になること(アインシュタイン 1909)、プランクの放射則(プランク 1901)、光量子仮説(アインシュタイン 1905)、放射の占有数が 1/(e の x 乗 − 1) であること、ウィーンの式が占有数の小さい極限、レイリー=ジーンズの式が大きい極限であること、ゆらぎが kT² dU/dT で書けることは、いずれも既知である。二重性の解釈をしない──光が波と粒子の「両方である」とも「どちらかである」とも言わない。言うのは、二つの項の比が占有数の逆数だという計算だけである。光子の同一性を導かない──占有数の分布が幾何分布であることは前提として置き、ボース統計そのものは導かない。光子の相関を扱わない──二つの検出器の相関(ハンブリー・ブラウンとトゥイス)、光子の束化、実在の光源(レーザーのポアソン分布、熱光源)の統計は主題ではない。熱平衡の一つのモードだけを見る。値は例である──5778 K・2.725 K・300 K・6000 K と 550 nm・10 μm・1 μm・160.2 GHz は比べるための代表値で、実在の光源の精度を主張しない。既刊との関係:論文1005 は、アインシュタインの帳簿から誘導放出を外すと釣り合う放射がウィーンの式になることを示した──ここでは、そのウィーンの式のゆらぎが粒子の項だけで、波の項は誘導放出と同じ側に消えることを数える。論文41 は、流束が強制する指数と縞を決める位相が別の根であることを示した──ここでは、同じ一つの式の中で、粒子の項と波の項がどちらの側に勝つかを占有数一つで数える。論文1011 は、最初の光電子の待ち時間が光子の到着のポアソン過程で書けることを示した──ここでは、可視光の占有数が 0.010926 で、その到着が粒子の側にあることを数える。論文902 は、黒体放射の係数の 4π が k 空間の単位球の面積であることを示した──ここでは、モードの数は動かさず、一つのモードの中のゆらぎだけを見る。加えたのは、二つの項の比が占有数の逆数であることを三つの道で確かめたこと、等しくなる点が ln 2 で占有数 1 であることを二分法で出したこと、プランクの山と太陽の可視光と背景放射をこの比で並べたこと、ウィーンとレイリー=ジーンズを片側の項として数で置いたことである。 第一に、これが本稿の芯である。粒子の項を波の項で割った比は占有数の逆数で、二つが等しいのは占有数がちょうど 1 のときである──一つのモードの平均エネルギーを u、光量子のエネルギーを hν とすると、ゆらぎの二乗平均は hν·u+u² で、前の項÷後の項=hν/u は e の x 乗から 1 を引いた値(x=hν/kT)に等しい。これが 1 になる x を二分法で求めると 0.693147=ln 2 で、そのときの占有数は 1.000000、粒子の項の割合は 0.500000 だった(第2節・第3節)。 第二に、三つの道が同じ数を返す──x=0.5・ln 2・2.821439・5 の四点で、アインシュタインの式、占有数の幾何分布の分散を和で出した値、熱力学の kT² dU/dT を数値微分した値は、いずれも 3.917698089・2.000000000・0.067292322・0.006829673 で一致した(第2節)。 第三に、プランクの山は粒子の側にある──振動数あたりの山 x=2.821439 では粒子の項が波の項の 15.801016 倍、波長あたりの山 x=4.965114 では 142.324918 倍で、宇宙背景放射の山 160.2 GHz でも 15.800801 倍である(第4節)。 第四に、太陽の可視光はさらに粒子の側で、等しくなるのは赤外である──5778 K の 550 nm では x=4.527445、比は 91.521904 倍、占有数は 0.010926 で、二つが等しくなる波長は 3.592448 μm である。2.725 K の背景放射では 7.617307 mm(39.356749 GHz)、300 K の物体では 69.19 μm で等しくなる(第4節)。 第五に、ウィーンの式は粒子の項だけ、レイリー=ジーンズの式は波の項だけである──x=8 でウィーンの式のゆらぎはプランクの和と相対差 -6.708x10^-4、x=0.05 でレイリー=ジーンズの式のゆらぎは相対差 0.000208 で、それぞれ片側の項だけで和に近づく(第5節)。 第六に、分離子は、占有数が 1 より大きいか小さいか──一つのモードに平均して光量子が一個より多く入っていれば波の項が勝ち、一個より少なければ粒子の項が勝つ。光が波か粒子かではなく、モードがどれだけ埋まっているかで、ゆらぎの形が決まる(第6節)。 光のゆらぎは粒子の項と波の項の和で、二つが等しいのは占有数がちょうど 1 のときである。一つのモードのゆらぎ hν·u+u² で、粒子の項÷波の項は占有数の逆数に等しく、hν=kT ln 2 で 1 になる。プランクの山では粒子の項が 15.801016 倍、太陽の可視光では 91.521904 倍で、等しくなるのは太陽なら 3.592448 μm、背景放射なら 7.617307 mm の側である──ウィーンの式は粒子の項だけ、レイリー=ジーンズの式は波の項だけを持つ。分離子は、占有数が 1 より大きいか小さいかである。既刊との位置──論文1005 は誘導放出を外すとウィーンの式になることを、論文41 は流束の指数と縞の位相が別の根であることを、論文1011 は光電子の到着がポアソン過程で書けることを示した。ここでは、一つの式の二つの項の勝ち負けが占有数一つで決まることを数えた。正直に言えば──光が何であるかについては、何一つ言っていない。 作成にあたって:本稿の着想と内容は、著者自身の考察に基づくものです。文章の構成整理や英訳、数式の確認には AI(大規模言語モデル)の助力を得ました。最終的な内容の解釈や誤りがあれば、それらはすべて著者の責に帰します。お気づきの点があれば、ご教示いただければ幸いです。 キーワード:輻射のゆらぎ、光量子、プランクの放射則、ウィーンの式、レイリー=ジーンズの式、占有数。

View source

Similar papers

#small language model Dataset Open access Oct 2026

Socratic guiding questions in synthetic arithmetic data: matched LoRA runs (revision v2)

Supporting data, adapters, predictions and code for the article *Low-Cost LoRA Fine-Tuning of Small Language Models for Multi-Step Arithmetic Reasoning* by Jake O'Grady, Asena Isik Gürhan, Chee Fong Ting and Effirul Ramlan (University of Galway). We generated 20,000 GSM8K-derived arithmetic problems with step-by-step s...

O'Grady, Jake, Gürhan, Asena Isik, Chee, Fong Ting et al. · 465 citations
#computer vision Open access Jun 2016

Software Development in Startup Companies: The Greenfield Startup Model

The results are packaged in the Greenfield Startup Model (GSM), which explains the priority of startups to release the product as quickly as possible, and the need to shorten time-to-market, by speeding up the development through low-precision engineering activities.

Carmine Giardino, Nicolò Paternoster, M. Unterkalmsteiner et al. · 178 citations · ⚡14
#computer vision Open access Oct 2016

Software Startups - A Research Agenda

Software startup companies develop innovative, software-intensive products within limited timeframes and with few resources, searching for sustainable and scalable business models.

M. Unterkalmsteiner, P. Abrahamsson, Xiaofeng Wang et al. · 157 citations · ⚡17
#machine learning Review Open access Oct 2016

“Failures” to be celebrated: an analysis of major pivots of software startups

This study conducts a case survey study based on the secondary data of the major pivots happened in 49 software startups, and demonstrates that customer need pivot is the most common among all pivot types.

Sohaib Shahid Bajwa, Xiaofeng Wang, Anh Nguyen-Duc et al. · 127 citations · ⚡15
#computer vision Review Open access May 2015

A survey study on major technical barriers affecting the decision to adopt cloud services

The comparison of adopter and non-adopter sample reveals three potential adoption inhibitor, security, data privacy, and portability, which underlines the importance of the technical and security perspectives for research investigating the adoption of technology.

Nattakarn Phaphoom, Xiaofeng Wang, S. Samuel et al. · 111 citations · ⚡8
#computer vision Open access Feb 2018

Lean Internal Startups for Software Product Innovation in Large Companies: Enablers and Inhibitors

This study investigates how Lean internal startup facilitates software product innovation in large companies and identifies its enablers and inhibitors, and shows the potential of the method-in-action framework to investigate the Lean startup approach in non-startup context.

Henry Edison, Nina M. Smørsgård, Xiaofeng Wang et al. · 78 citations · ⚡6

Related blog posts

Microsoft Research Blog Oct 12, 2026

Simulating student learning for better AI tutoring

StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simulating student learning for better AI tutoring appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.