Next Article in Journal
Imaging Velocimetry Measurements for Entropy Production in a Rotational Magnetic Stirring Tank and Parallel Channel Flow
Next Article in Special Issue
Gibbs’ Paradox in the Light of Newton’s Notion of State
Previous Article in Journal
A Universal Framework for Analysis of Self-Replication Phenomena
Previous Article in Special Issue
Entropy Calculation of Reversible Mixing of Ideal Gases Shows Absence of Gibbs Paradox
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

Thermodynamics of the System of Distinguishable Particles

Department of Physics, National Changhua University of Education, Taiwan
Entropy 2009, 11(3), 326-333; https://doi.org/10.3390/e11030326
Submission received: 16 April 2009 / Accepted: 25 June 2009 / Published: 29 June 2009
(This article belongs to the Special Issue Gibbs Paradox and Its Resolutions)

Abstract

:
The issue of the thermodynamics of a system of distinguishable particles is discussed in this paper. In constructing the statistical mechanics of distinguishable particles from the definition of Boltzmann entropy, it is found that the entropy is not extensive. The inextensivity leads to the so-called Gibbs paradox in which the mixing entropy of two identical classical gases increases. Lots of literature from different points of view were created to resolve the paradox. In this paper, starting from the Boltzmann entropy, we present the thermodynamics of the system of distinguishable particles. A straightforward way to get the corrected Boltzmann counting is shown. The corrected Boltzmann counting factor can be justified in classical statistical mechanics.

1. Introduction

The concepts of distinguishable and indistinguishable particles is important in Statistical Mechanics as their corresponding entropies are different. The entropy in statistical mechanics is defined in terms of the logarithm of the number of the accessible microstates in the phase space. The definition of the entropy is called Boltzmann entropy in which it is adopted in popular textbooks [1,2,3,4]. The microstates numbers for distinguishable and indistinguishable particles are certainly different and then their corresponding Boltzmann entropies are different, too. However, it is not trivial to know whether the distinguishability property may lead to different physical results. For example, the Gibbs paradox [6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31] presents one of the cases.
In its simplest case of the Gibbs paradox, the entropy of two identical volume of (distinguishable) ideal gas increases after mixture in which it violates our intuition. Consider two subsystems of equal volumes V and particle numbers N are separated by a wall, the Boltzmann entropy of one system is
S = k ln V N
Hence the total entropy S i of two systems is double,
S i = 2 k ln V N
in which the additivity of the Boltzmann entropy is assumed. Now the wall is removed, the gases of two systems mix. The entropy S f is
S f = k ln ( 2 V ) 2 N
Thus the mixing entropy reads
Δ S = S f - S i = 2 N k ln 2
which is positive meaning that the mixing process is irreversible according to the second law of thermodynamics. Here we also assume the Boltzmann entropy is equivalent to the usual thermodynamical entropy (that is, the entropy identified in thermal properties).
The essence of the paradox is in fact that the entropy of the distinguishable particles is not extensive. The entropy of the mixture is not equal to the sum of their partition. To resolve the Gibbs paradox, one introduces the indistinguishable particles in which a permutation factor 1 / N ! (N is the total particle number) is included in the total microstate number to overcome the overcounting [6]. Hence, the entropy before mixing in Eq. (2) should be modified as
S i = 2 k ln V N N !
and similarly, the entropy after mixing in Eq. (3) reads
S f = k ln ( 2 V ) 2 N ( 2 N ) !
The mixing entropy is
Δ S = 2 N k ln 2 - k ln ( 2 N ) ! ( N ! ) 2 = O ( ln N )
In the thermodynamical limit N , the mixing entropy per particle vanishes, and hence the Gibbs paradox is resolved.
The concept of indistinguishable particles is trivial in Quantum Mechanics. The identical particles in Quantum Mechanics are indistinguishable particles. Strictly speaking, one cannot distinguish two identical particles after their collision. At equilibrium, the number of microstates of the whole system is described in terms of the number representation (or the second quantized representation). The resolution of the Gibbs paradox is then straightforward in views of the quantum nature of identical particles.
However, we have also similar case in Classical Mechanics, for example, colloids. The colloidal particles (giant molecules) of size up to microns are distinguishable since its behavior should be governed by classical mechanics. In such a classical system, the entropy is certainly not extensive [2,3] and the Gibbs paradox appears. Traditionally, to avoid the Gibbs paradox, colloidal particles are treated as indistinguishable particles [34] without any explicit reason. There is still a puzzle even though this kind of treatment turns out to be correct.
In this paper, we present the thermodynamics of the system of distinguishable particles, starting from the definition of entropy for the distinguishable particles. The presentation can easily show how the “reduced" entropy [32] instead of the original entropy determines the thermodynamical behaviors.
Before introducing our treatment for distinguishable particles, we classify the system of particles into three categories,
  • Distinguishable particles of the same species
  • Indistinguishable particles (certainly of the same species)
  • Particles of different species (certainly distinguishable)
by their phase spaces. Suppose we call the phase space of the system of N distinguishable particles of the same species (the first category) be Γ N . The phase space of N indistinguishable particles (the second category) is then Γ N / S N with the permutation group S N . The permutation group S N is used to eliminate the overcounting microstate numbers from Γ N .
For the third category, each particle belongs to its particular species. The corresponding phase space of N particles of different species is Γ 1 Γ 1 Γ 1 (total N direct products).
The thermodynamics for the second and third categories were already well discussed in the textbook [2].

2. Thermodynamics of Distinguishable Particles

To study the thermodynamical variables of the distinguishable particle system, we consider two subsystems of particle number N 1 , N 2 , volumes V 1 , V 2 , energies E 1 , E 2 , respectively, in which their particles, volumes, and energies are allowed to be exchanged.
Before two subsystems in contact, the phase space is Γ N 1 Γ N 2 . After contact, the phase space becomes Γ N 1 + N 2 which is larger than Γ N 1 Γ N 2 by ( N 1 + N 2 ) ! / ( N 1 ! N 2 ! ) times. This number is the number of way to select N 1 particles and N 2 particles from the total N 1 + N 2 particles. That is,
Γ N 1 + N 2 = ( Γ N 1 Γ N 2 ) ( Γ N 1 Γ N 2 )
in which there are total ( N 1 + N 2 ) ! / ( N 1 ! N 2 ! ) copies (or configurational degeneracy). This degeneracy was pointed out by Penrose [5].
The total entropy as a whole becomes
S = S 1 ( N 1 , V 1 , E 1 ) + S 2 ( N 2 , V 2 , E 2 ) + k ln ( N 1 + N 2 ) ! N 1 ! N 2 !
S 1 and S 2 are the entropies of the subsystems separately. The last term arises from the configurational degeneracy. The existence of the additional term is due to the nonextensive property of the system. The distinguishability implies nonextensivity of the entropy [3] in which the total system entropy is not the naive addition of the subsystem entropies. Similar cases are also well known in other physical system [33].
Because of that, it is easily seen that the Gibbs paradox usually mentioned in the textbook [2] vanishes when we look at the phase space of the system of distinguishable particles carefully.
When two subsystems are in equilibrium, the total entropy attains its maximum, then we have
S = S 1 ( N 1 , V 1 , E 1 ) + S 2 ( N 2 , V 2 , E 2 ) + S 0 ( N 1 , N 2 )
with S 0 = k ln ( ( N 1 + N 2 ) ! / ( N 1 ! N 2 ! ) ) from Eq. (9). Under the constraint of energy E 1 + E 2 = E , volume V 1 + V 2 = V , and particle number N 1 + N 2 = N , the equilibrium attains when the entropy becomes extremum such that
S E 1 = S 1 E 1 - S 2 E 2 = 0
S V 1 = S 1 V 1 - S 2 V 2 = 0
S N 1 = S 1 N 1 - S 2 N 2 + S 0 N 1 = 0
We can then obtain, for the system of distinguishable particles, the expression of the temperature T as
1 T = S 1 E 1 V 1 , N 1 = S 2 E 2 V 2 , N 2
the pressure P as
P T = S 1 V 1 E 1 , N 1 = S 2 V 2 E 2 , N 2
By noticing that
S 0 N 1 = - N 1 ( k ln N 1 ! ) + N 2 ( k ln N 2 ! )
we can express the chemical potential μ as
μ T = - ( S 1 - k ln N 1 ! ) N 1 E 1 , V 1 = - ( S 2 - k ln N 2 ! ) N 2 E 2 , V 2
Now we introduce the “reduced" entropy [32]
S red = S - k ln N !
such that the above expressions for thermodynamical variables
1 T = S red E V , N
P T = S red V E , N
μ T = - S red N E , V
are re-written in the usual way of thermodynamics for indistinguishable particles. For the system of distinguishable particles, the “reduced" entropy S red instead of the original entropy S defines the thermodynamical variables of the system. Our presentation is somehow a straightforward way to justify the “reduced" entropy adopted for the classical (distinguishable) systems in condensed matter physics [34].
For the case of indistinguishable particles defined in the phase space Γ N / S N , its entropy is simply equivalent to the “reduced" entropy mentioned above. Although the original entropies for both the distinguishable and indistinguishable particles are different, the entropies governing their corresponding thermodynamics are still the same. Thermodynamics cannot tell the distinguishability of the system.
Suppose our system is N 1 , V 1 , E 1 in contact with the reservoir of N 2 , V 2 , E 2 in which N 1 N 2 , V 1 V 2 , E 1 E 2 , the entropy of the reservoir can be expanded into Taylor’s series around N , V , E such that
S = S 1 ( N 1 , V 1 , E 1 ) + S 2 ( N 2 , V 2 , E 2 ) + S 0 ( N 1 , N 2 ) = S 1 red ( N 1 , V 1 , E 1 ) + S 2 red ( N 2 , V 2 , E 2 ) + k ln N ! (22) S 2 red ( N 2 , V 2 , E 2 ) + k ln N !
in which the system entropy S 1 is neglected assuming S 1 S 2 . The reduced entropy S 2 red can be analyzed by Taylor’s series expansion around N , V , E up to first order, that is,
S 2 red ( N 2 , V 2 , E 2 ) = S 2 red ( N - N 1 , V - V 1 , E - E 1 ) = S 2 red ( N , V , E ) - N 1 S 2 red N 2 E , V , N 2 = N - V 1 S 2 red V 2 E , N , V 2 = V - E 1 S 2 red E 2 E 2 = E (23) = S 2 red ( N , V , E ) - 1 T ( - μ N 1 + E 1 + V 1 P )
with the temperature of the reservoir T.
The total entropy becomes
S = S 2 red ( N , V , E ) - 1 T ( - μ N 1 + E 1 + V 1 P ) + k ln N ! (24) = - 1 T ( - μ N 1 + E 1 + P V 1 ) + C
where C is a constant independent of N 1 , E 1 , V 1 . The probability p ( N 1 , V 1 , E 1 ) of the system is proportional to the corresponding phase space volume. The phase space we are now considering is Γ N 1 + N 2 , and hence from Eq. (8) there are exactly N ! / ( N 1 ! N 2 ! ) copies of the state characterized by N 1 , V 1 , E 1 . We have
p ( N 1 , V 1 , E 1 ) N ! N 1 ! N 2 ! Ω 1 ( N 1 , V 1 , E 1 ) 1 N 1 ! Ω 1 ( N 1 , V 1 , E 1 ) (25) 1 N 1 ! exp ( μ N 1 - E 1 - P V 1 k T )
under the condition that N 1 N 2 .
The probability distribution allows us to formulate the Grand Canonical ensemble in which our system interacts with a particle-energy reservoir. The grand partition function
Ξ ( μ , V , T ) = N r = 0 s 1 N r ! e μ N r / k T e - E s / k T (26) = N r = 0 e μ N r / k T Z N r ( V , T )
with the partition function
Z N r ( V , T ) = 1 N r ! s e - E s / k T
of the corrected Boltzmann counting due to the factor 1 / N r ! .

3. Conclusion

In summary, although the phase spaces of the system of distinguishable particles is different from that of indistinguishable one, their thermodynamics are in fact equivalent. It also implies that the corrected Boltzmann counting factor can be justified in classical statistical mechanics.

Acknowledgements

The author thanks Prof. Pik-Yin Lai and Prof. Shin-Chuan Gou for their helpful comments and discussions. Very useful comment from the reviewer is highly acknowledged. Support from the National Science Council of the Republic of China is acknowledged.

References

  1. Landau, L.D.; Lifshitz, E.M. Statistical Physics; Pergamon Press: Oxford, UK, 1980. [Google Scholar]
  2. Pathria, R.K. Statistical Mechanics; Butterworth-Heinemann: Boston, MA, USA, 1996. [Google Scholar]
  3. Reif, F. Fundamentals of Statistical and Thermal Physics; McGraw-Hill: New York, NY, USA, 1965; pp. 243–254. [Google Scholar]
  4. Recently Swendsen presented the different point of view on the defintion of entropy. Please refer to the references [7,8,9].
  5. Penrose, O. Foundations of Statistical Mechanics; Pergamon Press: Oxford, UK, 1970; pp. 163–164. [Google Scholar]
  6. Gibbs, J.W. Elementary Principles of Statistical Mechanics; Dover, 1960; pp. 206–207. [Google Scholar] After the paradox pointed out by Gibbs, there are a huge number of papers on the discussion and resolutions of the paradox. Please refer to the references [7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31].
  7. Swendsen, R.H. Statistical mechanics of classical systems of distinguishable particles. J. Statist. Phys. 2002, 107, 1143–1165. [Google Scholar] [CrossRef]
  8. Swendsen, R.H. Statistical mechanics of colloids and Boltzmann’s definition of the entropy. Am. J. Phys. 2006, 74, 187–190. [Google Scholar] [CrossRef]
  9. Swendsen, R.H. Gibbs Paradox and the Definition of Entropy. Entropy 2008, 10, 15–18. [Google Scholar] [CrossRef]
  10. Klein, M.J. Note on a problem concerning the Gibbs paradox. Am. J. Phys. 1958, 26, 80–81. [Google Scholar] [CrossRef]
  11. Noyes, R.M. Entropy of mixing of interconvertible species. Am. J. Phys. 1961, 34, 1983–1985. [Google Scholar]
  12. Fong, P. Semipermeable membrane and the Gibbs maradox. Am. J. Phys. 1964, 32, 170–171. [Google Scholar] [CrossRef]
  13. Boyer, T.H. Sharpening Bridgman’s resolution of the Gibbs paradox. Am. J. Phys. 1970, 38, 771–773. [Google Scholar] [CrossRef]
  14. Alexanian, M. Gibbs paradox in particle physics. Phys. Rev. D 1971, 4, 2432–2439. [Google Scholar] [CrossRef]
  15. Casper, B.M.; Frerer, S. “Gibbs Paradox" paradox. Am. J. Phys. 1973, 41, 509. [Google Scholar] [CrossRef]
  16. Landsberg, P.T.; Tranah, D. The Gibbs paradox and quantum gases. Am. J. Phys. 1978, 46, 228–230. [Google Scholar] [CrossRef]
  17. Lesk, A.M. On the Gibbs paradox: what does indistinguishability really mean? J. Phys. A 1980, 13, L111–L114. [Google Scholar] [CrossRef]
  18. Grandy, W.T. Indistinguishability, symmetrisation and maximum entropy. Eur. Phys. J. 1981, 2, 86. [Google Scholar] [CrossRef]
  19. Keesing, R.G. Lost work and the entropy of mixing. Eur. J. Phys. 1986, 7, 266–268. [Google Scholar] [CrossRef]
  20. Ben-Naim, A. Is mixing a thermodynamical process? Am. J. Phys. 1987, 55, 725. [Google Scholar] [CrossRef]
  21. Biedenharn, L.C.; Truini, P.; van Dam, H. A remark on possible violations of the Pauli principle. J. Phys. A 1989, 22, L67–L71. [Google Scholar] [CrossRef]
  22. Jaynes, E.T. The Gibbs paradox. In Maximum Entropy and Bayesian Methods; Smith, C.R., Erickson, G.J., Neudorfer, P.O., Eds.; Kluwer Academic: Dordrecht, Netherlands, 1992; pp. 1–22. [Google Scholar]
  23. Warren, P. Statistical mechanics of polydispersity, combinatorial entropy, and the Gibbs paradox. Phys. Rev. Lett. 1998, 80, 1369–1372. [Google Scholar] [CrossRef]
  24. Lin, S.-K. Gibbs paradox of entropy of mixing. Elect. J. Theor. Chem. 1996, 1, 135–150. [Google Scholar] [CrossRef]
  25. Tseng, C.Y.; Caticha, A. Yet another resolution of the Gibbs paradox: an information theory approach. In Bayesian Inference and Maximum Entropy Methods in Science and Engineering; Fry, R.L., Ed.; AIP Conference Proceedings; 2002. [Google Scholar]
  26. Allahverdyan, A.E.; Nieuwenhuizen, T.M. Explanation of the Gibbs paradox within the framework of quantum thermodynamics. Phys. Rev. E 2006, 73, 066119. [Google Scholar] [CrossRef] [PubMed]
  27. Ben-Naim, A. The entropy of mixing and assimilation: an information-theoretical perspective. Am. J. Phys. 2006, 74, 1126–1135. [Google Scholar] [CrossRef]
  28. Ben-Naim, A. On the So-Called Gibbs Paradox and on the Real Paradox. Entropy 2007, 9, 132–136. [Google Scholar] [CrossRef]
  29. Kiefer, C.; Kolland, G. Gibbs paradox and black-hole entropy. General Relativity and Gravitation 2008, 40, 1327–1339. [Google Scholar] [CrossRef]
  30. Maslov, V.P. Solution of the Gibbs paradox in the framework of classical mechanics (Statistical Physics) and crystallization of the gas C60. Math. Notes 2008, 83, 716–722. [Google Scholar] [CrossRef]
  31. Lin, S.-K. Gibbs Paradox and the Concepts of Information, Symmetry, Similarity and Their Relationship. Entropy 2008, 10, 1–5. [Google Scholar] [CrossRef]
  32. The “reduced entropy" is S - kln N! in which S is the original entropy and N is the number of particles in the system. We adopted the terminology because it is the same as the entropy of the system assuming that the particles in the system are indistinguishable. “Reduced" here means the reduction of the distinguishablility degrees of freedom.
  33. For example, if we consider the surface energy of the system, then the energy of the system is not proportional to its mass (nonextensive). When two subsystems are in contact, their total energy are not equal to the sum of the original two energies. Another example is the system with long-ranged electrostatic interaction.
  34. Chaikin, P.M.; Lubensky, T.C. Principles of Condensed Matter Physics; Cambridge University Press: Cambridge, UK, 1995. [Google Scholar]

Share and Cite

MDPI and ACS Style

Cheng, C.-H. Thermodynamics of the System of Distinguishable Particles. Entropy 2009, 11, 326-333. https://doi.org/10.3390/e11030326

AMA Style

Cheng C-H. Thermodynamics of the System of Distinguishable Particles. Entropy. 2009; 11(3):326-333. https://doi.org/10.3390/e11030326

Chicago/Turabian Style

Cheng, Chi-Ho. 2009. "Thermodynamics of the System of Distinguishable Particles" Entropy 11, no. 3: 326-333. https://doi.org/10.3390/e11030326

Article Metrics

Back to TopTop