Solutions for Series of Exponential Equations in Terms of Lambert-W Function and Fundamental Constants

Series of exponential equations in the form of were solved graphically, numerically and analytically. The analytical solution was derived in terms of Lambert-W function. A general numerical solution for any y is found in terms of n or in base y. A solution is close to the fine structure constant. The equation which provided the solution as the fine structure constant was derived in terms of the fundamental constants.

• Exponential Equation
• Lambert-W Function
• Fine Structure Constant
• Logarithmic Equation
• Numerical Analysis
• Fundamental Constants

1. Introduction

Exponential equations are widely used in natural and social sciences. In this paper, we considered series of exponential equations and solved them graphically, numerically, and analytically in terms of Lambert-W function. One equation connected to the fine structure constant, was derived in terms of the fundamental constants and led to a new equation. The Lambert-W function for real variables is defined by the equation W ( x ) exp [ W ( x ) ] = x [1] [2] [3] [4] and it has applications in Planks spectral distribution law [5] [6] , QCD renormalization [7] , solar cells [8] , bio-chemical kinetics [9] , optics [10] , population growth and water movement in soil [11] .

Considering the series of exponential equations defined by the following equation

x = y n y ( x y n + 1 ) (1.1)

where x, y, n are real variables.

Taking log_y on both sides of the Equation (1.1)

log y x = n + x y n + 1 (1.2)

Converting the Equation (1.2) to natural logarithm

ln x ln y = x y n + 1 + n (1.3)

The trivial solution of the Equations (1.1) to (1.3) is

x = y n + 1 (1.4)

In this paper, we are focusing on the non-trivial solutions.

For n = 2, 1, 0, −1, −2, the Equations (1.1) and (1.3) become:

x = y 2 y x y 3 or ln x ln y = x y 3 + 2 (1.5)

x = y y x y 2 or ln x ln y = x y 2 + 1 (1.6)

x = y x y or ln x ln y = x y or x y = y x (1.7)

x = y − 1 y x or ln x ln y = x − 1 (1.8)

x = y − 2 y x y or ln x ln y = x y − 2 (1.9)

2. Graphical Solutions

If y =10, the Equations (1.5) to (1.9) become

x = 10 2 × 10 x 10 3 (2.1)

x = 10 × 10 x 10 2 (2.2)

x = 10 x 10 (2.3)

x = 10 − 1 × 10 x (2.4)

x = 10 − 2 × 10 10 x (2.5)

The curves and the straight line to obtain the graphical solutions of the Equations (2.1) to (2.5) are shown in Figure 1. The intercepts of the curves and the straight line indicate the solutions.

The intersecting points of 0.1, 1, 10, 100 and 1000 are the trivial solutions and the intersecting points at around 0.0137, 0.137, 1.37, 13.7 and 137 are the non-trivial solutions.

The non-trivial solutions imply the following equations:

10 0.1371 = 1.371 (2.6)

ln 1.371 1.371 = ln 10 10 = 0.2302 (2.7)

10 1.371 = 1.371 10 = 23.5 (2.8)

3. Numerical Solutions

Higher precision non-trivial numerical solutions were obtained for the series of equations x = y n y ( x y n + 1 ) using the iterative technique for n = 2, 1.5, 1, 0.5, 0, −0.5, −1, −2 and 1 ≤ y ≤ 15 (Table 1). The iterations do not converge on non-trivial solutions for y < e, and solutions in this range were obtained by trial and error.

The solutions in Table 1 for n = −2, −1, 0, 0.5, 1, 2 are plotted as x vs y with x axis in log scale (Figure 2). Sharp turning points in the plots are observed for y values in the range of 1 to 2.

4. Analytical Solution

Consider the Equation (1.3)

ln x ln y = x y n + 1 + n

Table 1

Non-trivial numerical solutions for the series of equations x = y n y ( x y n + 1 )

y	Solutions x for different n values
	−2	−1	−0.5	0	0.5	1	1.5	2
1.1	36	39.6	41.53	43.56	45.68	47.92	50.25	52.71
1.3	7.41	9.63	10.99	12.53	14.28	16.28	18.56	21.17
1.5	3.29	4.94	6.05	7.41	9.07	11.11	13.61	16.67
2	1	2	2.83	4	5.66	8	11.31	16
e	0.366937	1.00000	1.644494	2.718282	4.46877	7.3890	12.13412	20.0855
3	0.275339	0.82601	1.430704	2.478052	4.292113	7.4341	12.87634	22.3024
4	0.125	0.500000	1	2.000000	4	8.0000	16	32.0000
5	0.070597	0.352984	0.789297	1.764922	3.946485	8.8246	19.73243	44.1230
6	0.045118	0.270707	0.663095	1.624244	3.978569	9.7454	23.87141	58.4727
7	0.031227	0.218591	0.578339	1.530140	4.04837	10.7109	28.33859	74.9768
8	0.022852	0.182813	0.517072	1.462501	4.136579	11.7000	33.09263	93.6000
9	0.017424	0.156820	0.470461	1.411382	4.234145	12.7024	38.10731	114.321
10	0.013713	0.137129	0.43364	1.371299	4.336395	13.7129	43.36395	137.129
11	0.011066	0.121721	0.403704	1.338936	4.440749	14.7282	48.84823	162.011
12	0.009113	0.109353	0.37881	1.312235	4.545715	15.7468	54.54858	188.961
13	0.007632	0.099215	0.357724	1.289792	4.650411	16.7672	60.45534	217.974
14	0.006483	0.090760	0.339593	1.270640	4.7543	17.7889	66.56021	249.045
15	0.005574	0.083606	0.323804	1.254088	4.857064	18.8113	72.85595	282.169

Let

t = − ln x

Then (1.3) becomes

− t ln y = e − t y n + 1 + n

( t + n ln y ) e t = − ln y y n + 1

( t + n ln y ) e t + n ln y = − e n ln y ln y y n + 1

( t + n ln y ) e t + n ln y = − ln y y

( t + n ln y ) = W ( − ln y y )

Substituting −lnx for t

( − ln x + n ln y ) = W ( − ln y y )

Using the Equation (1.3)

− x ln y y n + 1 = W ( − ln y y )

Hence the solution to Equation (1.3) is

x = W ( − ln y y ) ( − ln y y n + 1 ) = y n W ( − ln y y ) ( − ln y y ) (3.1)

If n = 0, the Equation (1.1) x = y n y ( x y n + 1 ) becomes Equation (1.7) x = y ( x y ) .

Using the solution in the Equation (3.1), the analytical solution in terms of Lambert-W function is

x = W ( − ln ( y ) y ) ( − ln ( y ) y ) (3.2) [12] [13]

In Equation (3.2), if y = e , x = W ( − 1 e ) ( − 1 e ) .

But W ( 1 e ) = − 1 [6] .

Hence x = e , the result in Table 1.

If n =0 and y =2 in Equation (3.2), the solutions in Table 1 and Equation (1.7) gives

2 4 = 4 2 and ln 2 2 = ln 4 4 = − 0.346. (3.3)

Equation (3.1) gives

4 = W ( − ln ( 2 ) 2 ) ( − ln ( 2 ) 2 ) = W ( − 0.346 ) − 0.346

2 = W ( − ln ( 4 ) 4 ) ( − ln ( 4 ) 4 ) = W ( − 0.346 ) − 0.346

W ( − 0.346 ) is double valued with−0.693 and −1.386.

If we substitute the solutions for n = 0 and y = 10 from Table 1 to Equation (3.2);

W ( − ln ( 1.37129 ) 1.37129 ) = 10 ( − ln ( 1.37129 ) 1.37129 ) (3.4)

W ( − 0.2302 ) = − 2.302 (3.5)

Since x and y are symmetric in Equation (1.7)

W ( − ln ( 10 ) 10 ) = 1.371289 ( − ln ( 10 ) 10 ) (3.6)

W ( − 0.2302 ) = − 0.3157 (3.7)

The W(x) has two real values for − 1 / e ≤ x < 0 [1] .

If n = −1, the Equation (1.1) x = y n y ( x y n + 1 ) becomes Equation (1.8) x = y − 1 y x or x y = y x .

Using the solution in the Equation (3.1), the analytical solution in terms of the Lambert-W function is

x = W ( − ln y y ) − ln y (3.8) [12]

If y = e , x = W ( − 1 e ) − ln e .

But W ( − 1 e ) = − 1 , Hence x = 1, the result in Table 1.

In Table 1, for any value of n y = e , x = e × e n , the trivial and nontrivial solutions coincide.

y = 10 , x = 1.371289 × 10 n

Using the solution in Equation (3.1), for any y the solution x can be written as

x ( n = 0 ) × y n (3.9)

Plots of lnx vs n shown in Figure 3 are linear as expected from Equation (3.1).

The lnx vs n lines for different y values are crossing near the point (0.5, 1.4). This indicates the solutions for n = 0.5 have little dependency on y for y ≥ e. This is also evident in the numerical results for n = 0.5 in Table 1 and in the plot of x = y 0.5 y x 1.5 in Figure 2.

5. Solutions x in Base y

The solutions x in Table 1 can be written in base y, (x_y) to indicate the general pattern.

For any valued of n, x_y can be written as

x y ( n = 0 ) × 10 n (5.1)

For n = 2, the solutions written in base y, x_y shown in Table 2.

For y > 11, the x_y are written using the hex notation.

There is a sharp change in the value of the x_y at y = 4.

For n = 2, plot x_y vs y, for 5 ≤ y ≤ 11 is shown in Figure 4.

6. Connection to the Fine Structure Constant

In Equation (1.1), when n = 2 and y = 10, the equation becomes x = 10 2 × 10 x 10 3 and the solution is

Table 2

Solutions x in base y(x y )

y	x written in base y for n = 2 xy
2	10000.000
e	1000.000
3	211.022
4	200.000
5	134.030
6	134.250
7	134.656
8	135.463
9	136.281
10	137.129
11	138.014
12	138.B60
13	139.C67
14	13B.090
15	13C.280
16	13D.590
17	13E.9A0
18	13F.F00

x = 10 2 W ( − ln 10 10 ) ( − ln 10 10 ) = 137.129 (6.1)

The solution 137.129 is close to the inverse of the fine structure constant 137.036 [14] - [21] which is dimensionless.

The inverse of the fine structure constant α − 1 is given by the expression

α − 1 = 4 π ε o ℏ c e 2 = 137.036 (6.2)

where;

ℏ = 1.0545718 × 10 − 34   J ⋅ s ,reduced Planck constant;

c = 2.99792458 × 10 8   m ⋅ s − 1 , speed of light in vacuum;

ε 0 = 8.854187817 × 10 − 12   F ⋅ m − 1 , electric constant;

e = 1.6021766208 × 10 − 19   C , elementary charge;

α − 1 , dimensionless constant [22] .

In a recent publication Eaves [23] suggested an equation relating G and α;

α q 2 8 π 2 G m e 2 ≈ exp ( 2 3 α ) (6.3)

where;

G = 6.67408 × 10 − 11   m 3 ⋅ kg − 1 ⋅ s − 2 , gravitational constant;

m e = 9.10938356 × 10 − 31   kg , electron mass.

q 2 = e 2 4 π ε 0

By substituting the expression for α in Equation (6.3) we get

e 4 32 π 3 ε 0 2 G m e 2 ℏ c ≈ exp ( 2 3 α ) (6.4)

Using Equation (6.2), the Equation (6.4) becomes

α − 1 ≈ 64 π 4 ε 0 3 ℏ 2 c 2 G m e 2 e 6 exp ( α − 1 1.5 ) (6.5)

Substituting numerical values for the pre-exponent,

α − 1 ≈ 1.59947 × 10 − 27 exp ( α − 1 1.5 ) (6.6)

α − 1 ≈ 1.59947 × 10 − 27 × 10 ( α − 1 3.4538 ) (6.7)

By taking the power of (1/289.5) on both sides of the Equation (6.8) and writing the equation for α − 1 yields

α − 1 ≈ 106.6 × 10 ( α − 1 1000 ) (6.8)

The Equation (6.8) is approximately the same as the equation x = 10 2 × 10 x 10 3 . The only difference is the 10² is 106.6 in Equation (6.8). But the Equation (6.8) based on the Equation (6.3) is only an approximate equation.

The value 1.59947 × 10 − 27 in Equation (6.6) is approximately equal to the α G 1 / 1.5 , α G defined by Jentschura [20] .

α G = G m e m p ℏ c = 3.21 × 10 − 42 (6.9)

Hence the Equation (6.7) can be written as

α − 1 ≈ α G 1 / 1.5 exp ( α − 1 1.5 ) (6.10)

7. Conclusions

An equation in the form of x = y n y ( x y n + 1 ) was solved graphically, numerically and analytically.

The plots of numerical solution x vs y indicate sharp turning points for y values in-between 1 to 2.

The analytical solution was found in terms of Lambert-W function as

x = y n W ( − ln y y ) ( − ln y y )

The numerical solutions can be written as x ( n = 0 ) × y n .

The numerical solutions can also be written in base y as x y ( n = 0 ) × 10 n . For y ≥ 5 x y ( n = 0 ) is a universal number approximately equal to 1.37.

If n = 2 and y = 10, the solution x = 10 2 W ( − ln 10 10 ) ( − ln 10 10 ) = 137.129 (rounded) is close to the inverse of the fine structure constant value, 137.036.

The equation x = 10 2 × 10 x 10 3 which gives the solution close to the fine structure constant can be derived from the equation α q 2 8 π 2 G m e 2 ≈ exp ( 2 3 α ) suggested by Eaves.

The derivation resulted in an equation α − 1 ≈ α G 1 / 1.5 exp ( α − 1 1.5 ) .

References

1. Corless, R.M., et al. (1996) On the LambertW Function. Advances in Computational Mathematics, 5, 329-359. https://doi.org/10.1007/BF02124750

2. Dence, T.P. (2013) A Brief Look into the Lambert W Function. Applied Mathematics, 4, 887. https://doi.org/10.4236/am.2013.46122

3. Kalman, D. (2001) A Generalized Logarithm for Exponential-Linear Equations. The College Mathematics Journal, 32, 2-14.https://doi.org/10.1080/07468342.2001.11921844

4. Fukushima, T. (2013) Precise and Fast Computation of Lambert W-Functions without Transcendental Function Evaluations. Journal of Computational and Applied Mathematics, 244, 77-89. https://doi.org/10.1016/j.cam.2012.11.021

5. Valluri, S.R., et al. (2009) The Lambert W Function and Quantum Statistics. Journal of Mathematical Physics, 50, Article Id: 102103. https://doi.org/10.1063/1.3230482

6. Valluri, S.R., Jeffrey, D.J. and Corless, R.M. (2000) Some Applications of the Lambert W Function to Physics. Canadian Journal of Physics, 78, 823-831.

7. Scott, T.C., Mann, R. and Martinez Ii, R.E. (2006) General Relativity and Quantum Mechanics: Towards a Generalization of the Lambert W Function A Generalization of the Lambert W Function. Applicable Algebra in Engineering, Communication and Computing, 17, 41-47. https://doi.org/10.1007/s00200-006-0196-1

8. Jain, A. and Kapoor, A. (2004) Exact Analytical Solutions of the Parameters of Real Solar Cells Using Lambert W-Function. Solar Energy Materials and Solar Cells, 81, 269-277. https://doi.org/10.1016/j.solmat.2003.11.018

9. Golicnik, M. (2012) On the Lambert W Function and Its Utility in Biochemical Kinetics. Biochemical Engineering Journal, 63, 116-123. https://doi.org/10.1016/j.bej.2012.01.010

10. Kitis, G. and Vlachos, N. (2013) General Semi-Analytical Expressions for TL, OSL and Other Luminescence Stimulation Modes Derived from the OTOR Model Using the Lambert W-Function. Radiation Measurements, 48, 47-54. https://doi.org/10.1016/j.radmeas.2012.09.006

11. Barry, D., et al. (1993) A Class of Exact Solutions for Richards’ Equation. Journal of Hydrology, 142, 29-46. https://doi.org/10.1016/0022-1694(93)90003-R

12. Weisstein, E.W. (2002) Lambert W-Function.

13. Gnanarajan, S. (2017) Solutions of the Exponential Equation yx/y = x or lnx/x = lny/y and Fine Structure Constant. Journal of Applied Mathematics and Physics, 5, 386. https://doi.org/10.4236/jamp.2017.52034

14. Kinoshita, T. (1996) The Fine Structure Constant. Reports on Progress in Physics, 59, 1459. https://doi.org/10.1088/0034-4885/59/11/003

15. Dirac, P.A. (1937) The Cosmological Constants. Nature, 139, 323. https://doi.org/10.1038/139323a0

16. Bouchendira, R., et al. (2011) New Determination of the Fine Structure Constant and Test of the Quantum Electrodynamics. Physical Review Letters, 106, Article ID: 080801. https://doi.org/10.1103/PhysRevLett.106.080801

17. Hammer, E. (2006) Physical and Mathematical Meaning of the Alpha Constant, Einstein’s Equation, and Planck Dimensions. Industry Applications Conference, Tampa, 8-12 October 2006. https://doi.org/10.1109/IAS.2006.256848

18. Kragh H. ,et al. (2003)Magic Number: A Partial History of the Fine-Structure Constant Archive for History of Exact Sciences 57, 395-431.

19. Sandvik, H.B., Barrow, J.D. and Magueijo, J. (2002) A Simple Cosmology with a Varying Fine Structure Constant. Physical Review Letters, 88, Article ID: 031302. https://doi.org/10.1103/PhysRevLett.88.031302

20. Jentschura, U.D. (2014) Fine-Structure Constant for Gravitational and Scalar Interactions. Physical Review A, 90, Article ID: 022112. https://doi.org/10.1103/PhysRevA.90.022112

21. Jentschura, U.D. and Nándori, I. (2014) Attempts at a Determination of the Fine-Structure Constant from First Principles: A Brief Historical Overview. The European Physical Journal H, 39, 591-613. https://doi.org/10.1140/epjh/e2014-50044-7

22. Mohr, P.J., Newell, D.B. and Taylor, B.N. (2016) CODATA Recommended Values of the Fundamental Physical Constants: 2014. Journal of Physical and Chemical Reference Data, 45, Article ID: 043102. https://doi.org/10.1063/1.4954402

23. Eaves, L. (2018) A Model to Inter-Relate the Values of the Quantum Electrodynamic, Gravitational and Cosmological Constants.