Comment on “Specific heat of an ideal Bose gas above the Bose condensation temperature,” [Am. J. Phys. 72 (9), 1193–1194 (2004)]

Frank Wang · 2026

Read Comment on “Specific heat of an ideal Bose gas above the Bose condensation temperature,” [Am. J. Phys. 72 (9), 1193–1194 (2004)] by Frank Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-18180.txt.

Introduction

Wang revisits Albert Einstein's 1925 paper on Bose-Einstein condensation through its English translation. He walks readers through Einstein's calculations for specific heat above the condensation temperature. Along the way he corrects some numerical errors and compares Einstein's formula with a different one published in 2004. A short history of how Einstein's theory gained acceptance rounds out the comment.

Catalog record

CREATORFrank Wang
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE12011
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.18180

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Comment on “Specific heat of an ideal Bose gas above the Bose condensation temperature,” [Am. J. Phys. 72 (9), 1193–1194 (2004)] Frank Wang Abstract In 2004, I presented a derivation for an approximate formula for the specific heat of an ideal Bose gas above the Bose-Einstein condensation temperature in this Journal . fw2004 I was trying to produce a graph of the specific heat resembling the Greek letter \Lambda , but failed to find an explicit formula in textbooks I consulted with back then. In 2025, Princeton University Press published The Essential Einstein , ee2025 a collection of Albert Einstein’s works including an English translation of Albert Einstein’s groundbreaking 1925 paper, einstein1925 and from it I learned that Einstein himself set a procedure for the formula I was searching for. In this Comment I will guide the readers to execute the calculations that Einstein outlined, correct some numerical errors in the paper, and compare his formula with mine. The history of the acceptance of Einstein’s theory will be summarized. In 1924, Einstein published “Quantum Theory of the Monoatomic Ideal Gas” (Ref. ee2025 , Chapter 24) which laid the foundation for quantum statistical mechanics. Einstein’s interest was triggered by the request to translate and publish a paper (previously rejected by the Philosophical Magazine ) by the then unknown Indian physicist Satyendranath Bose. Bose used an innovative way to count photons and obtained the Planck distribution; Einstein generalized Bose’s counting method for identical particles, and established what is now known as the Bose-Einstein distribution. In 1925, Einstein wrote a second paper with the same title (Ref. ee2025 , Chapter 25) and predicted the phenomenon of “Das gesättigte ideale Gas” (the saturated ideal gas) which is now referred to as a Bose-Einstein condensate. He ended the paper with the equation of state of the unsaturated ideal gas, see Figure 1 , which is the focus of this note. Einstein used E for the internal energy and the lower case n for the total number of bosons, and he wrote (Eq. (22) in Ref. ee2025 on page 353) \frac{E}{n}=\frac{3}{2}\kappa T\frac{\sum_{\tau}\tau^{-5/2}\lambda^{\tau}}{\sum_{\tau}\tau^{-3/2}\lambda^{\tau}}, (1) where he used \kappa for the Boltzmann constant, T for temperature, and \lambda is the fugacity e^{\mu/\kappa T} (where \mu is the chemical potential), which is a real number between 0 and 1. One can read Einstein’s original papers or Ref. fw2004 for this result; his internal energy is equivalent to Eq. (17) in Ref. fw2004 ; see the Appendix. Einstein denoted the denominator in Eq. ( 1 ) by y(\lambda) and the numerator by z(\lambda) . His y in his Eq. (18c) shown in Figure 1 is essentially a temperature parameter and can be written more compactly as y=\zeta(3/2)\left(\frac{T_{c}}{T}\right)^{3/2}, (2) where \zeta(3/2) is the Riemann zeta function evaluated at 3/2 and T_{c} is the critical temperature for Bose-Einstein condensation; see Ref. fw2004 Eq. (11). The value of \zeta(3/2) is 2.612375\ldots but Einstein used a crude value of 2.615 for it throughout his paper. (The value of \zeta(5/2) is 1.341487\ldots but Einstein used 1.348 in his Eqs. (40) and (41), and in Eq. (40) he or the typist missed a factor of 3/2 for the internal energy.) Figure 1: From The Essential Einstein ee2025 . Einstein sought to express z as a function of y , and he credited his assistant Jakob Grommer for the solution. For sufficiently small y one can write a Taylor expansion in y , and the coefficients can be obtained from repeated differentiation. In the present case the differentiation d^{\nu}z/dy^{\nu} involves parametric equations z(\lambda) and y(\lambda) , but the procedure is a standard topic in a modern-day calculus course. Einstein’s Eqs. (43) and (44) can be implemented using a computer algebra system such as Maple or Mathematica, which excels in this kind of repetitive and tedious process. z=y-\frac{1}{2!}\frac{1}{2\sqrt{2}}y^{2}-\frac{1}{3!}\left(\frac{4}{3\sqrt{3}}-\frac{3}{4}\right)y^{3}-\frac{1}{4!}\left(\frac{9}{4}+\frac{15\sqrt{2}}{8}-2\sqrt{6}\right)y^{4}+\ldots, (3) or z=y-0.1768\,y^{2}-0.0033\,y^{3}-\,0.00011y^{4}+\ldots. ( 3 ′ ) Einstein’s quartic term in his expansion z=y-0.1768\,y^{2}-0.0034\,y^{3}-0.0005\,y^{4} shown in Figure 1 was inaccurate; it is conceivable that he or his assistant committed an error while performing the laborious manual calculation. He introduced the notation F(y)=z/y , and with it the internal energy per particle is 3/2\,\kappa TF(y) , see his Eq. (19c). The specific heat is the derivative of the internal energy with respect to temperature, and it involves the following derivative rules. \frac{d}{dT}\left[TF(y)\right]=F(y)+T\frac{dF(y)}{dT}=F(y)+T\frac{dF(y)}{dy}\frac{dy}{dT}. (4) Because y is proportional to T^{-3/2} , T\,dy/dT=-3/2\,y . The specific heat per mole above T_{c} based on Einstein’s approach (see c_{v} after his Eq. (18c) in Figure 1 ) after correcting the numerical error in his z is c_{v}=\frac{3}{2}R(F(y)-\frac{3}{2}yF^{\prime}(y))=\frac{3}{2}R\left[1+0.231\left(\frac{T_{c}}{T}\right)^{3/2}+0.045\left(\frac{T_{c}}{T}\right)^{3}+0.0069\left(\frac{T_{c}}{T}\right)^{9/2}\right], (5) where R is the ideal gas constant and we use Eq. ( 2 ) for y after the differentiation. For a comparison, the specific heat above T_{c} in Ref. fw2004 is reproduced here. c_{v}=R\left\{\left[\frac{9}{2}\frac{\zeta(5/2)}{\zeta(3/2)}-\frac{3}{8}\frac{\zeta(3/2)^{2}}{\pi}\right]+\left[\frac{9}{4}\frac{\zeta(5/2)}{\zeta(3/2)}-\frac{3}{8}\frac{\zeta(3/2)^{2}}{\pi}\right]\left(\frac{T_{c}}{T}\right)^{3/2}+\left[\frac{3}{4}\frac{\zeta(3/2)^{2}}{\pi}-\frac{3\zeta(5/2)}{\zeta(3/2)}\right]\left(\frac{T_{c}}{T}\right)^{3}\right\}, (6) or c_{v}=\frac{3}{2}R\left[0.997+0.227\left(\frac{T_{c}}{T}\right)^{3/2}+0.057\left(\frac{T_{c}}{T}\right)^{3}\right]. ( 6 ′ ) Figure 2: Specific heat of an ideal Bose-Einstein gas; Eq. ( 7 ) for 0\leq T/T_{c}\leq 1 and Eq. ( 6 ) for T/T_{c}\geq 1 . The specific heat below T_{c} is well known; see Eq. (1) in Ref. fw2004 . c_{v}=\frac{15}{4}R\left(\frac{T}{T_{c}}\right)^{3/2}\frac{\zeta(5/2)}{\zeta(3/2)}=1.926\,R\left(\frac{T}{T_{c}}\right)^{3/2}. (7) With it the specific heat of an ideal Bose-Einstein gas is shown in Figure 2 . In this plot, it is visually indistinguishable between Einstein’s Eq. ( 5 ) and my Eq. ( 6 ) for the specific heat above T_{c} , but the two formulas are conceptually different. Einstein made the expansion around y=0 , which is the limit T\rightarrow\infty , while I made the expansion around T=T_{c} . Einstein’s specific heat tends to 3/2\,R as T tends to infinity, while my specific heat and the discontinuity of the first derivatives with respect to temperature (\partial c_{v}/\partial T) at T_{c} are exact in terms of \zeta(3/2) , \zeta(5/2) , and \pi . Coincidentally, Einstein’s high-temperature formula works fine near the critical temperature, and my critical-temperature formula is satisfactory at high temperature. After Einstein’s 1925 publication, “the degeneracy of the Bose-Einstein gas has rather got the reputation of having only a purely imaginary existence” london1938a and scientists overlooked this work for years. In 1938, P. Kapitza and independently J. F. Allen and D. Misener published results of the superfluidity of liquid helium. The same year Fritz London wrote a letter to Nature london1938a with a plot of the specific heat similar to Figure 2 . The resemblance between the \Lambda shape and the measured specific heat convinced most physicists about a direct link between Einstein’s theoretic work and experimental observables. In a 1939 essay written for Harper’s Magazine titled “The Usefulness of Useless Knowledge” flexner1939 by Abraham Flexner, the then director of the Institute for Advanced Study in Princeton, he highlighted Einstein’s quantum statistical research as an example that unexpected applications might emerge from seemingly “useless” studies. He quoted the following from Science : “The stature of Professor Albert Einstein’s genius reached new heights when it was disclosed that the learned mathematical physicist developed mathematics fifteen years ago which are now helping to solve the mysteries of the amazing fluidity of helium near the absolute zero of temperature scale.” In London’s letter to Nature , he wrote c_{v}=\frac{3}{2}R\left[1+0.231\left(\frac{T_{c}}{T}\right)^{3/2}+0.046\left(\frac{T_{c}}{T}\right)^{3}+\ldots\right] (8) for T\geq T_{c} . Comparing this with Eq. ( 5 ), London seemed to have noticed the numerical error in Einstein’s 1925 paper and

Read the complete text