SciELO - Scientific Electronic Library Online

 
vol.68 issue6Dynamics of the capillary rise in tilted Taylor-Hauksbee cellsRadiation from a dipole perpendicular to the interface between two planar semi-infinite magnetoelectric media author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • Have no similar articlesSimilars in SciELO

Share


Revista mexicana de física

Print version ISSN 0035-001X

Rev. mex. fis. vol.68 n.6 México Nov./Dec. 2022  Epub July 31, 2023

https://doi.org/10.31349/revmexfis.68.060602 

Research

Fluid Dynamics

The q-deformed heat equation and q-deformed diffusion equation with q-translation symmetry

W. Sang Chunga  * 

H. Hassanabadib  ** 

J. Křižc  *** 

a Department of Physics and Research Institute of Natural Science, College of Natural Science, Gyeongsang National University, Jinju 660-701, Korea.

b Faculty of physics, Shahrood University of Technology, Shahrood, Iran.

c Department of Physics, University of Hradec Králové, Rokitanského 62, 500 03 Hradec Králové, Czechia.


Abstract

In this paper we consider the discrete heat equation with a certain non-uniform space interval which is related to q-addition appearing in the non-extensive entropy theory. By taking the continuous limit, we obtain the q-deformed heat equation. Similarly, we obtain the solution of the q-deformed diffusion equation.

Keywords: q-deformed; q-translation

1 Introduction

Heat equation governs how heat diffuses or transfers through a region, which was first introduced by Fourier [1] in 1822. In one dimension, this equation take the form,

tu(x,t)=κx2u(x,t), (1)

where u(x, t) is the temperature at position x at time t and κ is thermal diffusivity.

In this paper we are to find a deformed heat equation. To do so we need the discrete version of heat equation where space is discrete but time is continuous. Discrete physics have been studied in various fields [2-15]. If we consider discrete positions denoted by

xn=na,nZ, (2)

we have the discrete heat equation,

tu(xn,t)=κΔx2u(xn,t), (3)

where finite difference operators are defined as

Δxu(xn,t)=u(xn+1,t)-u(xn,t)a . (4)

If we take the limit a0 in Eq. (4), we have Eq. (1). From Eq. (2), we know that

xn+1-xn=a, (5)

which implies that the uniform space interval guarantees the heat equation of the form (1). In other words, if we consider a non-uniform discrete position, we will obtain another form of heat equation.

In this paper we consider the discrete heat equation with a certain non-uniform space interval which is related to q-addition or q-subtraction appearing in the non-extensive entropy theory [16-18]. By taking the continuous limit, we obtain the q-deformed heat equation. Similarly, we derive the q-deformed diffusion equation. This paper is organized as follows: In Sec. 2 we discuss the q-deformed heat equation. In Sec. 3 we discuss the solution of q-deformed heat equation. In Sec. 4 we discuss cooling of a rod from a constant initial temperature. In Sec. 5 we discuss the q-deformed diffusion equation.

2 q-deformed heat equation

In this section we discuss the q-deformed heat equation based on the the q-addition and q-subtraction appearing in the non-extensive thermodynamics [16-18]. As is different from the non-extensive thermodynamics, we introduce the parameter q so that it may have a dimension of inverse length. In the non-extensive thermodynamics, the parameter q is dimensionless. Thus, in the q-deformed heat equation, q can be regarded as 1/ξ where ξ denotes a length scale.

Now let us introduce the discrete position with non-uniform interval where the distance between adjacent positions are given by

xn+1qxn=a, (6)

or

xn+1=xnqa, (7)

where the q-addition and q-subtraction [16-18] are defined as

aqb=a+b+qab, (8)

aqb=a-b1+qb. (9)

As is different from the uniform lattice, the non-uniform lattice consisting discrete points obeying Eq. (6) can be regarded as an example of the non-homogeneous medium in the continuous limit (a0). We think that the discrete positions defined by the different pseudo addition (deformation of the ordinary addition) can give another examples of the non-homogeneous medium in the continuous limit. For example, in Ref. [19], the α-addition was introduced to describe the non-homogeneous medium where anomalous diffusion arose.

The Eq. (6) gives the relation

xn+1=(1+qa)xn+a (10)

Solving Eq. (10) we get

xn=1q([1+qa]n-1), (11)

When q > 0 we have > 0 we have

limnxn= (12)

and

limn-xn=-1q  . (13)

When q > 0 we have < 0 we get

limnxn=1|q|, (14)

and

limn-xn=-. (15)

In this case we demand |q|a<1. The discrete position is not symmetric for x0 = 0. Indeed, we have

x-n=-xn(1+qa)n,n1. (16)

For the discrete positions obeying Eq. (6), the difference operator becomes

Δx:qu(xn,t)=u(xn+1,t)-u(xn,t)xn+1qxn=(1+qxn)×u(xn+1,t)-u(xn,t)xn+1-xn. (17)

Thus, in the continuum limit, we get

Δx:qu(xn,t)Dxq=(1+qx)ux. (18)

Here we know that the q-derivative Dxq remains invariant under the q-translation xxδx. Recently, quantum theory with q-translation invariance was constructed in [20]. Using Eq. (18), we obtain the q-deformed heat equation with q-translation symmetry in the form,

tu(x,t)=κDxq2u(x,t). (19)

3 Solution of q-deformed heat equation

Consider a rod of length L with the initial condition

u(x,0)=f(x), (20)

and the boundary condition

u(0,t)=u(L,t)=0. (21)

We look for a solution of the form

u(x,t)=X(x)T(t). (22)

Inserting Eq. (22) into Eq. (19) we get

1κTdTdt=1X(Dxq)2X=-λ,λ>0. (23)

Thus, we have

T(t)=e-κλt, (24)

and

X(x)=Acosλqln(1+qx)+Bsinλqln(1+qx), (25)

From the boundary function, we have A = 0 and

sinλqln(1+qL)=0, (26)

which gives

λ=λn=qnπln(1+qL),n=1,2, (27)

Thus, the general solution of q-deformed wave equation is

u(x,t)=n=1Bnsinnπln(1+qx)ln(1+qL)×exp-κq2n2π2t(ln(1+qL))2. (28)

Now let us apply the initial condition. Then we have

f(x)=n=1Bnsinnπln(1+qx)ln(1+qL). (29)

If we use the orthogonality relation

0Lsinnπln(1+qx)ln(1+qL)sinmπln(1+qx)ln(1+qL)×dx1+qx=ln(1+qL)2qδnm, (30)

we have

Bn=2qln(1+qL)0Lf(x)×sinnπln(1+qx)ln(1+qL)dx1+qx. (31)

Here we solved the q-heat equation in a closed form. Our method is to introduce the q-lattice as an example of the non-homogeneous medium, which is not related to the numerical solution methods based on adaptive grids [21-24] because we obtained the exact solution.

4 Cooling of a rod from a constant initial temperature

Suppose the initial temperature distribution f(x) in the rod is constant, i.e. f(x) = u0. Now let us consider the case of L = 1, κ= 1. Then we have

Bn=-2u0nπ((-1)n-1) (32)

Thus, we have

u(x,t)=4u0πn=112n-1sin(2n-1)πln(1+qx)ln(1+q)×exp-q2(2n-1)2π2t(ln(1+q))2. (33)

In this case, the ratio of the first and second terms in Eq. (33) is

| second term|| first term |=13e-8q2π2t(ln(1+q))2sin3πln(1+q)ln(1+q)sinπln(1+qx)ln(1+q), (34)

e-8q2π2t(ln(1+q))2, (35)

e-8forttq, (36)

where we used

|sinnt|n|sint|, (37)

and

tq=(ln(1+q))2q2π2. (38)

Thus, the first term dominates the sum of the rest of the terms, and hence

u(x,t)4u0πsinπln(1+qx)ln(1+q)×exp-q2π2t(ln(1+q))2. (39)

4.1 Spatial temperature profiles

Now let us consider fixed time. Here we consider the time t = tq. Then we have

u(x,t)4u0πe-1sinπln(1+qx)ln(1+q). (40)

This has the maxima at x = x0 where

x0=1+q-1q. (41)

Thus, center of a rod is not a line of symmetry unless q = 0. Fig. 1 shows the plot of u versus x with u0 = 1 for q = 0 (Red), q = 0.2 (Brown), and q = -0.2 (Gray). We know that the position for maximum of u is smaller than 1/2 for q > 0 while it is larger than 1/2 for q < 0.

Figure 1 Plot of u versus x with u0 = 1 for q = 0 (Red), q = 0.2 (Brown), and q = -0.2 (Gray). 

4.2 Temperature profiles in time

Setting x = x0 in the approximate solution, we have

u(x,t)4u0πexp-q2π2t(ln(1+q))2. (42)

Figure 2 shows the plot of u versus t with u0 = 1, x = x0 for q = 0 (Red), q = 0.2 (Brown), and q = -0.2 (Gray).

Figure 2 Plot of u versus t with u0 = 1, x = x0 for q = 0 (Red), q = 0.2 (Brown), and q = -0.2 (Gray). 

5 q-deformed diffusion equation

The q-deformed diffusion equation has the same form as the 𝑞-deformed heat equation,

tu(x,t)=D(Dxq)2u(x,t), (43)

where u(x, t) denotes the concentration and D denotes diffusivity. Now let us impose the initial condition

u(x,0)=f(x). (44)

Now let us introduce the 𝑞-deformed Fourier transform as

Fux,t=Uw,t=12π-1qdx1+qxu(x,t)(1+qx)iwqq>012π-1qdx1+qxu(x,t)(1+qx)iwqq<0, (45)

and the inverse q-deformed Fourier transform

F-1(U(w,t))=u(x,t)=-dwU(w,t)(1+qx)-iwq. (46)

From the definition of q-deformed Fourier transform, we know

F((Dxq)nu(x,t))=(-iw)nU(w,t). (47)

Taking the q-deformed Fourier transform in Eq. (43) we get

U(w,t)t=-Dw2U(w,t), (48)

which is solved as

U(w,t)=c(w)e-Dw2t. (49)

Then we have

U(w,0)=c(w). (50)

Thus we get

c(w)=F(f(x)). (51)

Now let us set

g(x)=F-1(e-Dw2t), (52)

which gives

g(x)=πDtexp-14Dt1qln(1+qx)2. (53)

Then we have

U(w,t)=F(f(x))F(g(x)). (54)

From the convolution theorem we get

u(x,t)=12π-1/qdx1+qxf(s)g1qln1+qx1+qs(q>0)12π-1/|q|dx1+qxf(s)g1qln1+qx1+qs(q<0)    (55)

If we impose the initial condition

f(x)=δ(x), (56)

we have

u(x,t)=14πDtexp-14Dt1qln(1+qx)2. (57)

From Eq. (57), the expectation value of x are

E(x)=1qeq2Dt(e3q2Dt-1). (58)

For a small q, we get

E(x)3qDt. (59)

The variance is then given by

V(x)=-2q2e5q2Dt(cosh(q2Dt)+cosh(3q2Dt)-cosh(4q2Dt)-sinh(q2Dt)-1). (60)

For a small q, we get

V(x)2Dt+16q2(Dt)2. (61)

Figure 3 shows the plot of u(x, t) versus x with t = 1 and D = 1 for q = 0 (Pink), q = 0.2 (Brown) and q = -0.2 (Gray). We know that the graph is asymmetric unless q = 0. Thus Eq. (57) is the asymmetric normal distribution. Figure 4 shows the plot of u(x, t) versus x with q = 0.2 and D = 1 for t = 1 (Pink), t = 2 (Brown) and t = 3 (Gray). Figure 5 shows the plot of u(x, t) versus x with q = 0.2 and D = 1 for t = 1 (Pink), t = 2 (Brown) and t = 3 (Gray).

Figure 3 Plot of u(x, t) versus x with t = 1 and D = 1 for q = 0 (Pink), q = 0.2 (Brown) and q = -0.2 (Gray). 

Figure 4 Plot of u(x, t) versus x with q = 0.2 and D = 1 for t = 1 (Pink), t = 2 (Brown) and t = 3 (Gray). 

Figure 5 Plot of u(x, t) versus x with q = -0.2 and D = 1 for t = 1 (Pink), t = 2 (Brown) and t = 3 (Gray). 

6 Conclusion

In this paper we studied the q-deformed heat equation and q-deformed diffusion equation. From the fact that the ordinary heat equation was obtained by taking continuous limit in the discrete heat equation with a uniform space interval, we considered the discrete heat equation with a certain non-uniform space interval which was related to q-addition or q-subtraction appearing in the non-extensive thermodynamics. By taking the continuous limit, we obtained the q-deformed heat equation. We found that the q-deformed heat equation possessed the q-translation symmetry instead of the ordinary translation. We solved the q-deformed heat equation for a rod of length L. We discussed cooling of a rod from a constant initial temperature. We used the q-deformed Fourier transform to find the solution of the q-deformed diffusion equation. We found that the variance in x takes the form,

V(x)=-2q2e5q2Dt(cosh(q2Dt)+cosh(3q2Dt)-cosh(4q2Dt)-sinh(q2Dt)-1). (62)

For a small q, we obtained

V(x)2Dt+16q2(Dt)2. (63)

We found that the q-deformed diffusion process is asymmetric.

The q-addition and q-subtraction defined in the non-extensive thermodynamics was rarely used in the deformation of the space-time (q-deformed space time). The application to quantum mechanics was discussed in [20], application to mechanics was discussed in [25], and construction of q-lattice and q-Bloch theorem was discussed in [26].

Besides, we comment the connection of non-homogeneous media with the q-deformation of space briefly. It seems impossible to describe the general non-homogeneous media in an exact way without numerical study. Here, we adopted a special non-homogeneous media related to q-deformed space. Because asymmetry of the q-lattice, the graphs of temperature in the q-heat equation and concentration in the q-deformed equation became asymmetric, which had different feature for positive q and negative q, (See Fig. 1-5).

Finally, we compare the q-deformed diffusion equation with the diffusion model with the effective position dependent diffusion coefficient D(x) [27-29]. The diffusion equation with the effective position dependent diffusion coefficient D(x) was given in the form,

u(x,t)t=xw(x)D(x)xu(x,t)w(x), (64)

where w(x) denotes the variable cross section. Comparing Eq. (43) with Eq. (64), we know

w(x)=1+qx,D(x)=D(1+qx)2. (65)

Thus we know that the q-deformed diffusion equation is an example of the diffusion model with the effective position dependent diffusion coefficient.

Acknowledgements

The authors thank the referee for a thorough reading of our manuscript and constructive suggestions.

References

1. J. Fourier, Theorie Analytique de la Chaleur Firmin Didot (1822). (reissued by Cambridge University Press, 2009; ISBN 978-1-108-00180-9). [ Links ]

2. J. Cadzow, Int. J. Control 11 (1970) 393. [ Links ]

3. J. Logan, Aequat. Math. 9 (1973) 210. [ Links ]

4. T. Durt, B. Englert, I. Bengtsson, and K. Zyczkowski, International Journal of quantum information 8 (2010) 535. [ Links ]

5. W. Wootters and B. Fields, Annals of Physics 191 (1989) 363. [ Links ]

6. C. Miquel, J. Paz, and M. Saraceno, Phys. Rev. A 65 (2002) 062309. [ Links ]

7. J. Paz, Phys. Rev. A 65 (2002) 062311. [ Links ]

8. E. Davies and J. Lewis, Communications in Mathematical Physics 17 (1970) 239. [ Links ]

9. A. Barth and A. Lang, Stochastics. 84 (2012) 217. [ Links ]

10. I. Gyöngy, Potential Anal. 9 (1998) 1. [ Links ]

11. M. Sauer and W. Stannat, Math. Comp. 84 (2015) 743. [ Links ]

12. T. Shardlow, Numer. Funct. Anal. Optim. 20 (1999) 121. [ Links ]

13. Y. Yan, SIAMJ. Numer. Anal. 43 (2005) 1363. [ Links ]

14. C. Mueller and K. Lee, Amer. Math. Soc. 137 (2009) 1467. [ Links ]

15. M. Friesl, A. Slavḱ, P. Stehlḱ, Appl. Math. Lett. 37 (2014) 86. [ Links ]

16. C. Tsallis, J. Stat. Phys. 52 (1988) 479. [ Links ]

17. E. Borges, Physica A 340 (2004) 95. [ Links ]

18. L. Nivanen, A. Le Mehaute, Q.Wang, Rep. Math. Phys. 52 (2003) 437. [ Links ]

19. W. Chung and H. Hassanabadi, Mod. Phys. Lett. B 33 (2019) 1950368. [ Links ]

20. Won Sang Chung and H. Hassanabadi, Fortschr. Phys. 2019 1800111. [ Links ]

21. C. De Boor, Good approximation by splines with variable knots II in Lecture Notes in Mathematics 363, Springer-Verlag, New York, 1974, pp. 12-20. [ Links ]

22. E. Dorfi and L. Drury, J. Comput. Phys. 69 (1987) 175. [ Links ]

23. A. Dwyer, R. Kee and B. Sanders, AIAAJ 18 (1980) 1205. [ Links ]

24. D. Hawken, J. Gottlieb and J. Hanssen, J. Comput. Phys. 95 (1991) 254. [ Links ]

25. W. Chung and H. Hassanabadi, q-deformed mechanics from the discrete mechanics in q-lattice and Nöther theorem PREPRINT 2022. [ Links ]

26. W. Chung and H. Hassanabadi, Bloch theorem and tight binding model in q-Bravais lattice PREPRINT 2022. [ Links ]

27. P. Kalinay and J. Percus, J. Chem. Phys. 122 (2005) 204701. [ Links ]

28. R. Zwanzig, J. Stat. Phys. 9 (1973) 215. [ Links ]

29. D. Reguera and J. Rubi, Phys. Rev. E 64 (2001) 061106. [ Links ]

Received: October 18, 2021; Accepted: May 19, 2022

* E-mail: e-mail: mimip44@naver.com

** E-mail: h.hasanabadi@shahroodut.ac.ir

*** E-mail: jan.kriz@uhk.cz

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License