betaFeedback

WorksPhysics and Astronomy

An Accurate Zeroth-Order Perturbation Theory for Solving Power-Law Potentials within the Framework of the Asymptotic Iteration Method

Abstract

Spherically symmetric singular potentials have a wide range of applications, such as molecular and atomic physics. In this work, we present a zeroth-order perturbation theory for solving the Schrödinger equation within the framework of the asymptotic iteration method (AIM). We rely on exponential and hyperbolic potential ansatz substitutions, where the difference between the original and new potential is treated as a small perturbation. Each potential ansatz has a free parameter chosen small enough to give an accurate zeroth-order perturbative wave equation solution. This method avoids the oscillatory behavior observed in the asymptotic iteration solutions when applying the AIM to such problems. Finally, we illustrate our approach by solving a few problems, particularly potentials with quartic and sextic singularities. We compare our results with those obtained via other methods and find good agreement.

Introduction

Exact and numerical solutions of the Schrödinger equation are of both theoretical and experimental importance. Hamiltonians with spherically symmetric potentials, in particular, have found a wide range of applications in molecular and nuclear physics [1, 2, 3]. For example, Lennard-Jones’ potential (aka six-twelve potential) is used to model interactions in neutral molecules [4]. It is well-known that the space of exactly solvable solutions of the Schrodinger equation is relatively limited; hence there is always a need for numerical techniques to solve the problem under consideration. The Lagrange-mesh methods [5], methods of finite difference [6, 7], and the asymptotic iteration method (AIM) [8] are some of the numerical methods attributed to solving many eigenvalue problems.

In this work, we use AIM to study the power-law potentials with spherical symmetry. In practice, the application of AIM to such problems is limited due to the oscillatory behaviour in the numerical computations generated by AIM iterative process; therefore, convergence is not easily obtained in several cases [9, 10, 11].

An earlier proposal to overcome this problem was introduced in [12], where a perturbative approach is proposed and applied cautiously. A reason for this careful examination has to do with the computational demand of the perturbative approaches and the need for higher-order perturbative calculations. The approach was also highly sensitive to the asymptotic wave-function solution required to put the eigenvalue problem in the appropriate form for AIM. Furthermore, the convergence rate was highly sensitive to the initial seed point x0x_0 needed to utilize the iterative procedure.

In the present work, we introduced, in the next section, a less demanding approach where we formulate the problem in such a way that we obtain the results we are looking for within a reasonable accuracy and manageable computational time. In § 3, we initiate this technique using two different classes of potential ansatz. In Section 4, we provide applications to our method by solving three classes of problems, and finally, we conclude our work in Section 5.

Zeroth-order perturbation theory within the frame of the AIM

The radial Schrodinger equation under consideration is

[−12d2dr2+V(r)]ψ(r)=Eψ(r)(1)\left[-\frac{1}{2}\frac{d^2}{dr^2}+V(r)\right]\psi(r)=E\psi(r) \tag*{(1)}

where V(r)V(r) is the effective potential that is difficult to address using the straightforward application of AIM. Our starting point is to add and subtract an Ansatz potential V0(λ,r)V_0(\lambda,r) to the above equation, giving

[−12d2dr2+V0(λ,r)+Uλ(r)]ψ(r)=Eψ(r)(2)\left[-\frac{1}{2}\frac{d^2}{dr^2}+V_0(\lambda,r)+U_\lambda(r)\right]\psi(r)=E\psi(r) \tag*{(2)}

where

Uλ(r)=V(r)−V0(λ,r)(3)U_\lambda(r)=V(r)-V_0(\lambda,r) \tag*{(3)}

is a regular function at the origin (r=0r=0) and λ\lambda that is small enough so that Uλ(r)U_\lambda(r) can be treated as a perturbation that satisfy lim⁡λ→0Uλ(r)=0\lim_{\lambda\to0}U_\lambda(r)=0. AIM is then used to solve the Schrödinger equation

[−12d2dr2+V0(λ,r)]ψλ(0)(r)=Eλ(0)ψλ(0)(r)(4)\left[-\frac{1}{2}\frac{d^2}{dr^2}+V_0(\lambda,r)\right]\psi_\lambda^{(0)}(r)=E_\lambda^{(0)}\psi_\lambda^{(0)}(r) \tag*{(4)}

Usually a coordinate transformation y=y(r)y=y(r) and a wavefunction Ansatz ψλ(0)(y)=g(y)f(y)\psi_\lambda^{(0)}(y)=g(y)f(y) is used to map the above equation into a form suitable to AIM application

d2f(y)dy2=k0(y,Eλ(0))df(y)dy+s0(y,Eλ(0))f(y)(5)\frac{d^2f(y)}{dy^2}=k_0(y,E_\lambda^{(0)})\frac{df(y)}{dy}+s_0(y,E_\lambda^{(0)})f(y) \tag*{(5)}

where

k0(y,Eλ(0))=−2(1g(y)dg(y)dy+y′′2y′2)(6)k_0(y,E_\lambda^{(0)})=-2\left(\frac{1}{g(y)}\frac{dg(y)}{dy}+\frac{y''}{2y'^2}\right) \tag*{(6)}

and

s0(y,Eλ(0))=−y′′2y′2g(y)d2g(y)dy2−y′′y′2g(y)dg(y)dy+2V0(λ,y)−2Eλ(0)y′2.(7)s_0(y,E_\lambda^{(0)})=-\frac{y''^2}{y'^2g(y)}\frac{d^2g(y)}{dy^2}-\frac{y''}{y'^2g(y)}\frac{dg(y)}{dy}+\frac{2V_0(\lambda,y)-2E_\lambda^{(0)}}{y'^2}. \tag*{(7)}

The AIM sequences kn(y)k_n(y) and sn(y)s_n(y), n=1,2,⋯n=1,2,\cdots are generated recursively via

\begin{pmatrix} k_n(y) \\ s_n(y) \end{pmatrix} = \begin{pmatrix} \frac{d}{dy}+k_0 & \frac{1}{d} \\ s_0 & \frac{d}{dy} \end{pmatrix} \begin{pmatrix} k_{n-1}(y) \\ s_{n-1}(y) \end{pmatrix}. \tag*{(8)}\end{pmatrix}

Assume that the ratio sn(y)κn(y)\frac{s_n(y)}{\kappa_n(y)} converges asymptotically to some function η(y)\eta(y) for sufficiently large nn [8], the energy spectrum are calculated as the roots of the termination condition

Δn(y,E(0))=κn(y)sn−1(y)−κn−1(y)sn(y)(9)\Delta_n(y,E^{(0)})=\kappa_n(y)s_{n-1}(y)-\kappa_{n-1}(y)s_n(y) \tag*{(9)}

and the wave functions ψλ(0)(y)\psi_\lambda^{(0)}(y), see (4), are establish using [8]

f(y)=exp⁡(−∫0yη(t) dt).(10)f(y)=\exp\left(-\int_0^y\eta(t)\,\mathrm{d}t\right). \tag*{(10)}

In comparison with the perturbation theory in quantum mechanics [13], E(0)E^{(0)} is the zeroth-order perturbative solution of the original wave equation. We choose our perturbation parameter λ\lambda such that the lowest-order solution gives accurate results; this is possible by controlling the free parameter λ\lambda that comes with the potential V0(r)V_0(r). Higher-order corrections to the energy and wave function can be calculated via the well-known time-independent perturbation theory if needed. We may note the following two identities

lim⁡λ→0ψλ(0)(r)=ψ(r)(11)\lim_{\lambda\to0}\psi_\lambda^{(0)}(r)=\psi(r) \tag*{(11)}

and

lim⁡λ→0Eλ(0)=E.(12)\lim_{\lambda\to0}E_\lambda^{(0)}=E. \tag*{(12)}

Potential ansatz substitutions

Consider the following general form of the power-law potential

V(r)=a1r+a2r2+⋯+anrn,(13)V(r)=\frac{a_1}{r}+\frac{a_2}{r^2}+\cdots+\frac{a_n}{r^n}, \tag*{(13)}

we propose the following two Ansatz for V0(λ,r)V_0(\lambda,r)

V0(λ,r)=∑k=0nAk(eλr−1)−k(14)V_0(\lambda,r)=\sum_{k=0}^{n}A_k\left(e^{\lambda r}-1\right)^{-k} \tag*{(14)}

and

V0(λ,r)=∑k=0nBk(sinh⁡λr)−k(15)V_0(\lambda,r)=\sum_{k=0}^{n}B_k(\sinh\lambda r)^{-k} \tag*{(15)}

where the coefficients {Ak,Bk}\{A_k,B_k\} are chosen such that

Uλ(r)=V(r)−V0(λ,r)→0,as λ→0.(16)U_\lambda(r)=V(r)-V_0(\lambda,r)\to0,\quad\text{as }\lambda\to0. \tag*{(16)}

Generally speaking, λ\lambda is a free parameter chosen to be small such that Uλ(r)U_\lambda(r) is negligible compared to V0(λ,r)V_0(\lambda,r). We refer to the Ansatz in Eq. (14) by the ‘Exponential Ansatz’ (EA), while the Ansatz in Eq. (15) is referred to as the ‘Hyperbolic Ansatz (HA).

In Appendix § 7, we have listed the Ansatz choices (14)–(15) for power-low potentials with up to sextic singularities. In the next section, we apply those Ansatz to a variety of problems and find the corresponding discrete spectra using the AIM.

Applications

The Coulomb potential

One of the simplest and exactly solvable potentials is the Coulomb problem,

V(r)=−Qr+ℓ(ℓ+1)2r2(17)V(r)=-\frac{Q}{r}+\frac{\ell(\ell+1)}{2r^2} \tag*{(17)}

where QQ is the charge of the particle (e.g. electron) and ℓ=0,1,…\ell=0,1,\ldots is the angular momentum quantum number. The above potential corresponds to the following exact energy spectrum formula

Enℓ=−12(Qn+ℓ+1)2,n=0,1,2,…(18)E_{n\ell}=-\frac{1}{2}\left(\frac{Q}{n+\ell+1}\right)^2,\qquad n=0,1,2,\ldots \tag*{(18)}

Starting with the exponential Ansatz, we use the substitutions from Appendix § 7 for v1=1/rv_1=1/r, and v2=1/r2v_2=1/r^2. For the EA ansatz, use the change of variable y=1−2e−λry=1-2e^{-\lambda r} and we take g(y)=1−y2g(y)=1-y^2. For the resulting equation, Table 1 reports the numerical eigenvalues obtained using AIM. For these values, the seed (initial) value y0=0y_0=0 was used, and the perturbation parameter λ=0.001\lambda=0.001. Table 1 reflects the good agreement with the exact eigenvalues obtained using (18).

ℓ=0\ell= 0ℓ=1\ell= 1ℓ=2\ell= 2
ExactAIMExactAIMExactAIM
-0.5-0.5000000419-0.125-0.1250004167-0.0555555556-0.05555643056
-0.125-0.1250005000-0.0555555556-0.05555659722-0.03125-0.03125175000
-0.0555555556-0.05555668056-0.03125-0.03125191667-0.02-0.02000287500
-0.03125-0.03125200000-0.02-0.02000304167-0.0138888889-0.01389313889
-0.02-0.02000312500-0.0138888889-0.01389330556-0.0102040816-0.01020995663
-0.0138888889-0.01389338889-0.0102040816-0.01021012330-0.0078125-0.007820250000
-0.0102040816-0.01021020663-0.0078125-0.007820416667-0.0061728395-0.006182714506
-0.0078125-0.007820500000-0.0061728395-0.006182881173-0.005-0.005012250000

Table 1.

We may also employ the HA to find the energy spectrum for the same parameters chosen above, and we compared the results with those obtained via the EA ansatz as shown in Table 2. In our AIM calculations for the HA case, we applied the transformation y=e−λry=e^{-\lambda r} and g(y)=1g(y)=1, and we used the seed point y0=0.5y_0=0.5 with λ=0.001\lambda=0.001. We found that the two Ansatz are in good agreement with each other. Furthermore, we made a plot of the potential Ansatz for different values of the potential parameters and compared that with the exact potential as shown in Figure 1.

Figure 1. A plot of Coulomb potential and the Ansatz potentials for Q=1Q=1 and ℓ=0,1\ell=0,1.

ℓ=0\ell= 0ℓ=1\ell= 1ℓ=2\ell= 2
EAHAEAHAEAHA
-0.5000000419-0.4735505486-0.1250004167-0.1249991667-0.05555643056-0.05555380558
-0.1250005000-0.1249990000-0.05555659722-0.05555347228-0.03125175000-0.03124650025
-0.05555668056-0.05555330564-0.03125191667-0.03124616706-0.02000287500-0.01999425121
-0.03125200000-0.03124600047-0.02000304167-0.01999391823-0.01389313889-0.01388039293
-0.02000312500-0.01999375176-0.01389330556-0.01388006035-0.01020995663-0.01019234249
-0.01389338889-0.01387989409-0.01021012330-0.01019201059-0.007820250000-0.007797025166
-0.01021020663-0.01019184466-0.007820416667-0.007796694296-0.006182714506-0.006153141844
-0.007820500000-0.007796528896-0.006182881173-0.006152812476-0.005012250000-0.004975600141

Table 2.

Table 1: The energy spectrum of the Coulomb potential obtained via the exact well-known formula versus the results obtained via the Exponential Ansatz with Q=1Q = 1, y0=0y_0 = 0, λ=0.001\lambda= 0.001, and 400 iterations.

Table 2: Q=1Q = 1 Hyperbolic Ansatz (HA) vs Exponential Ansatz (EA). In EA, we used x0=0x_0 = 0, λ=0.001\lambda= 0.001, while in the HA, we took x0=0.5x_0 = 0.5, λ=0.001\lambda= 0.001

The spiked isotropic oscillator

Another interesting problem to apply our Ansatz method is the spiked oscillator with quartic singularity [14, 15, 16, 17, 18, 19, 20]

V(r)=12ω2r2+a22r4+ℓ(ℓ+1)2r2(19)V(r) = \frac{1}{2}\omega^2 r^2 + \frac{a^2}{2r^4} + \frac{\ell(\ell+ 1)}{2r^2} \tag*{(19)}

The potential Ansatz is chosen based on the transformations in Appendix § 7. However, for the quadratic term, we use the identity r2=1/(1/r2)r^2 = 1/(1/r^2), and then we use the corresponding substitution for 1/r21/r^2 from the appendix. For the hyperbolic ansatz, we took y=sinh⁡(λr)y = \sinh(\lambda r), g(y)=e−y2/2g(y) = e^{-y^2/2}, y0=1/10y_0 = 1/10 and λ=0.01\lambda= 0.01. We took the parameters ω=1\omega= 1 and a=0.5a = 0.5 and compared our results for the lowest ten energies obtained via the TRA [20] as shown in Table 3. We find good agreement between our results and those obtained via the TRA. For the EA, we took y=e−λr−1y = e^{-\lambda r} - 1, g(y)=1g(y) = 1, y0=1/15y_0 = 1/15 and λ=0.01\lambda= 0.01. A comparison between the results of both ansatzes is given in Table 4. We also plotted our Ansatz and compared them with the exact potential {(19)} as shown in Figure 2.

Figure 2. A plot of spiked isotropic oscillator potential and its Ansatz potentials for ω=1\omega= 1, a=0.5a = 0.5 and ℓ=0,1\ell= 0, 1.

ℓ=3\ell= 3ℓ=4\ell= 4ℓ=5\ell= 5ℓ=6\ell= 6
TRAAIMTRAAIMTRAAIMTRAAIM
4.5140936644.5145064375.5078958815.5084947066.5050381396.5058528527.5034917507.504555983
6.5201995266.5211434787.5107349047.5119863928.5065799018.5081461759.5044200529.506335250
8.5262017938.5278893929.5135588559.51566369010.5081181410.5106363111.5053473011.50831359
10.5321539510.5347488811.5163632311.5195268212.5096528812.5133232413.5062734913.51049099
12.5380473712.5417261413.5191609613.5235760114.5111841514.5162069615.5071986415.51286741
14.5438920714.5488249615.5219408215.5278114416.5127119716.5192874717.5081227317.51544282
16.5495529216.5560487817.5247077517.5322333418.5142363718.5225647619.5090457919.51821721
18.5552912218.5634007419.5274591019.5368418720.5157573720.5260388221.5099678121.52119054
20.5605435520.5708837421.5301978221.5416372322.5172749922.5297096423.5108887923.52436279
22.5658054522.5785004423.5329207523.5466195924.5187892524.5335772125.5118087425.52773392

Table 3. The first few lowest energies of the spiked oscillator Eq. (19) for ω=1\omega= 1, a=0.5a = 0.5 obtained via the TRA [20] versus the Hyperbolic Ansatz (HA). For the HA, we took y0=0.1y_0 = 0.1, λ=0.01\lambda= 0.01 and 400 iterations.

ℓ=3\ell= 3ℓ=4\ell= 4ℓ=5\ell= 5ℓ=6\ell= 6
EAHAEAHAEAHAEAHA
4.5138810424.5145064375.5075964585.5084947066.5046307566.5058528527.5029595997.504555983
6.5196856336.5211434787.5101091207.5119863928.5057966508.5081461759.5034623099.506335250
8.5252966158.5278893929.5125062559.51566369010.5068587910.5106363111.5038638211.50831359
10.5307188010.5347488811.5147881311.5195268212.5078172112.5133232413.5041641513.51049099
12.5359564812.5417261413.5169550013.5235760114.5086719314.5162069615.5043632915.51286741
14.5410135414.5488249615.5190071215.5278114416.5094229716.5192874717.5044612417.51544282
16.5458935216.5560487817.5209447217.5322333418.5100703718.5225647619.5044580119.51821721
18.5505996518.5634007419.5227682719.5368418720.5106141420.5260388221.5043537221.52119054
20.5551349120.5708837421.5244655621.5416372322.5110543122.5297096423.5041426423.52436279
22.5595020622.5785004423.5265919323.5466195924.5113909024.5335772125.5040832725.52773392

Table 4. Comparison between the HA and the EA lowest ten energies for the spiked oscillator with ω=1\omega= 1, a=0.5a = 0.5 and various values of ℓ\ell as indicated.

The power-law potential with sextic singularity

To show the power of our method, we choose a problem with sextic singularity,

V(r)=ℓ(ℓ+1)2r2−b2r4+12a4r6(20)V(r) = \frac{\ell(\ell+ 1)}{2r^2} - \frac{b^2}{r^4} + \frac{1}{2}\frac{a^4}{r^6} \tag*{(20)}

where the real parameters {a,b}\{a, b\} are chosen such that ∣b∣>∣a∣|b| > |a|. By starting with the exponential Ansatz, we find V0(λ,r)\mathcal{V}_0(\lambda,r) by replacing 1/r21/r^2, 1/r41/r^4, and 1/r61/r^6 in V(r)V(r) with v2v_2, v4v_4, and v6v_6 from Appendix § 7, respectively. Following the analysis in § 2, we choose a coordinate transformation y=1−2e−λry = 1 - 2e^{-\lambda r} and g(y)=(1−y)3(1+y)4g(y) = (1-y)^3(1+y)^4 to obtain the AIM second-order differential equation that matches the form of Eq. (5). As an illustration, we took a=2a = 2, b=7b = 7, considered several values of the angular momentum ℓ:1−6\ell: 1 - 6 and calculated the corresponding energy spectrum as reported in Table 5. To examine the efficiency of our approach, we compared the calculated eigenvalues with those obtained via the higher-order finite difference scheme (HOFD) [6] as shown in Table 5. In our AIM computations, we took y0=2/15y_0 = 2/15 and λ=0.5\lambda= 0.5. A good agreement between our results and the data from the HOFD has been observed. On the other hand, using the HA potential, we make a change of variable y=2tanh⁡2(λr)−1y = 2\tanh^2(\lambda r)-1 and g(y)=(1−y)10(1+y)10g(y) = (1-y)^{10}(1+y)^{10}. Solving Eq. (9) for E(0)E^{(0)} with the use of y0=0y_0 = 0 and λ=0.5\lambda= 0.5, we obtain the corresponding approximate energy spectrum of Eq. (20) as shown in Table 6. The accuracy of those approximate ansatz is also seen from their plots versus the exact potential as illustrated in Fig. ??.

ℓ=0\ell= 0ℓ=1\ell= 1ℓ=2\ell= 2
HOFDAIMHOFDAIMHOFDAIM
-198.053811725-198.0537328-194.485061786-194.4832736-187.382284129-187.3817611
-95.107391509-95.10726998-92.507287345-92.50381690-87.351046558-87.35019401
-37.503430893-37.50323491-35.790300903-35.78702065-32.419721196-32.41860615
-10.649435153-10.64906542-9.698773354-9.696562310-7.867299662-7.865513250
-1.563859663-1.562902864-1.193393381-1.190909350-0.540802068-0.5360973375
ℓ=3\ell= 3ℓ=4\ell= 4ℓ=5\ell= 5
HOFDAIMHOFDAIMHOFDAIM
-176.816074818-176.8143073-162.895376078-162.8935434-145.771664441-145.7696640
-79.728921145-79.72578334-69.782624776-69.77953504-57.714757854-57.71149979
-27.507836187-27.50448038-21.242775651-21.23900784-13.908937252-13.90422352
-5.305987630-5.302008650-2.281242651-2.274819916

Table 5. The energy spectrum of (20) for the parameters a=2a = 2, b=7b = 7, and different values of ℓ\ell using the AIM with the EA potential versus the results obtain via the HOFD. For the AIM, we took 400 iterations.

ℓ=0\ell= 0ℓ=0\ell= 0ℓ=1\ell= 1ℓ=1\ell= 1ℓ=2\ell= 2ℓ=2\ell= 2
HAEAHAEAHAEA
-198.0491077-198.0537328-194.4740844-194.4832736-187.3713222-187.3817611
-95.10036570-95.10726998-92.48865866-92.50381690-87.33340002-87.35019401
-37.49199053-37.50323491-35.76710592-35.78702065-32.39420310-32.41860615
-10.62838677-10.64906542-9.667043170-9.696562310-7.823876188-7.865513250
-1.518415325-1.562902864-1.129380489-1.190909350-0.4330104614-0.5360973375
ℓ=3\ell= 3ℓ=3\ell= 3ℓ=4\ell= 4ℓ=4\ell= 4ℓ=5\ell= 5ℓ=5\ell= 5
HAEAHAEAHAEA
-176.8021730-176.8143073-162.8781685-162.8935434-145.7473592-145.7696640
-79.70705001-79.72578334-69.75589780-69.77953504-57.67625301-57.71149979
-27.47327862-27.50448038-21.19529103-21.23900784-13.83791731-13.90422352
-5.239053277-5.302008650-2.172355430-2.274819916

Table 6. a=2a = 2, b=7b = 7 Hyperbolic Ansatz (HA) vs Exponential Ansatz (EA). In EA, we used x0=2/15x_0 = 2/15, λ=0.5\lambda= 0.5, while in the HA, we took x0=0x_0 = 0, λ=0.5\lambda= 0.5

Conclusion

Based on the zeroth order perturbation theory, we have proposed, in the AIM framework, two approximate schemes of the radial Schrödinger equation involving power-law potentials. We provided several examples to illustrate the efficiency of our approach. It was possible to study the challenging case of the supersingular sextic potential despite the failure of the traditional AIM due to the high instability in finding the root of the termination condition. Since we use a zeroth order perturbation method, our approach has a minimal computational cost compared to other techniques used to study such classes of singular potentials. The present work opens a door to study various eigenvalues in the AIM framework that were previously difficult to treat. We may emphasize that we have indirectly solved new classes of eigenvalue problems generated by the potential Ansatz and can be treated using AIM.

Acknowledgments

Partial financial support of this work under Grant No. GP249507 from the Natural Sciences and Engineering Research Council of Canada is gratefully acknowledged.

Potential Ansatz up to sextic singularities

The Hulthen type Ansatz

Let un=1rnu_n = \frac{1}{r^n} and the potential we are willing to study V(r)V(r) is a linear combination of such terms. The Ansatz potential V0(r)V_0(r) is obtained by replacing every unu_n with an alternative expansion vnv_n. Here, we list the first six expansions

1λv1=1eλr−1+12,\frac{1}{\lambda}v_1 = \frac{1}{e^{\lambda r}-1} + \frac{1}{2},
1λ2v2=(1eλr−1)2+1λr−512,\frac{1}{\lambda^2}v_2 = \left(\frac{1}{e^{\lambda r}-1}\right)^2 + \frac{1}{\lambda r} - \frac{5}{12},
1λ3v3=(1eλr−1)3+32(1eλr−1)2+121eλr−1,\frac{1}{\lambda^3}v_3 = \left(\frac{1}{e^{\lambda r}-1}\right)^3 + \frac{3}{2}\left(\frac{1}{e^{\lambda r}-1}\right)^2 + \frac{1}{2}\frac{1}{e^{\lambda r}-1},
1λ4v4=(1eλr−1)4+2(1eλr−1)3+76(1eλr−1)2+161eλr−1−1720,\frac{1}{\lambda^4}v_4 = \left(\frac{1}{e^{\lambda r}-1}\right)^4 + 2\left(\frac{1}{e^{\lambda r}-1}\right)^3 + \frac{7}{6}\left(\frac{1}{e^{\lambda r}-1}\right)^2 + \frac{1}{6}\frac{1}{e^{\lambda r}-1} - \frac{1}{720},
1λ5v5=(1eλr−1)5+52(1eλr−1)4+2512(1eλr−1)3+1524(1eλr−1)2+1241eλr−1,\frac{1}{\lambda^5}v_5 = \left(\frac{1}{e^{\lambda r}-1}\right)^5 + \frac{5}{2}\left(\frac{1}{e^{\lambda r}-1}\right)^4 + \frac{25}{12}\left(\frac{1}{e^{\lambda r}-1}\right)^3 + \frac{15}{24}\left(\frac{1}{e^{\lambda r}-1}\right)^2 + \frac{1}{24}\frac{1}{e^{\lambda r}-1},
1λ6v6=(1eλr−1)6+3(1eλr−1)5+134(1eλr−1)4+32(1eλr−1)3+31120(1eλr−1)2+11201eλr−1+130240.\begin{aligned} \frac{1}{\lambda^6}v_6 ={}& \left(\frac{1}{e^{\lambda r}-1}\right)^6 + 3\left(\frac{1}{e^{\lambda r}-1}\right)^5 + \frac{13}{4}\left(\frac{1}{e^{\lambda r}-1}\right)^4 + \frac{3}{2}\left(\frac{1}{e^{\lambda r}-1}\right)^3 \\ &+ \frac{31}{120}\left(\frac{1}{e^{\lambda r}-1}\right)^2 + \frac{1}{120}\frac{1}{e^{\lambda r}-1} + \frac{1}{30240}. \end{aligned}

The hyperbolic Ansatz

Similar to the previous section, we obtain an alternative potential Ansatz by replacing every unu_n by wnw_n, where we list the first six expansions below

1λw1=1sinh⁡(λr)1λ2w2=1sinh⁡2(λr)+131λ3w3=1sinh⁡3(λr)+121sinh⁡(λr)1λ4w4=1sinh⁡4(λr)+231sinh⁡2(λr)−1451λ5w5=1sinh⁡5(λr)+561sinh⁡3(λr)+5121sinh⁡(λr)−381sinh⁡(λr)1λ6w6=1sinh⁡6(λr)+1sinh⁡4(λr)+2151sinh⁡2(λr)+2945\begin{aligned} \frac{1}{\lambda}w_1 &= \frac{1}{\sinh(\lambda r)} \\ \frac{1}{\lambda^2}w_2 &= \frac{1}{\sinh^2(\lambda r)}+\frac{1}{3} \\ \frac{1}{\lambda^3}w_3 &= \frac{1}{\sinh^3(\lambda r)}+\frac{1}{2}\frac{1}{\sinh(\lambda r)} \\ \frac{1}{\lambda^4}w_4 &= \frac{1}{\sinh^4(\lambda r)}+\frac{2}{3}\frac{1}{\sinh^2(\lambda r)}-\frac{1}{45} \\ \frac{1}{\lambda^5}w_5 &= \frac{1}{\sinh^5(\lambda r)}+\frac{5}{6}\frac{1}{\sinh^3(\lambda r)}+\frac{5}{12}\frac{1}{\sinh(\lambda r)}-\frac{3}{8}\frac{1}{\sinh(\lambda r)} \\ \frac{1}{\lambda^6}w_6 &= \frac{1}{\sinh^6(\lambda r)}+\frac{1}{\sinh^4(\lambda r)}+\frac{2}{15}\frac{1}{\sinh^2(\lambda r)}+\frac{2}{945} \end{aligned}

References

  1. [1]Shi-Hai dong, Exact Solutions of the Two-Dimensional Schrödinger Equation with Certain Central Potentials. International Journal of Theoretical Physics 39, 1119 - 1128 (2000).DOI
  2. [2]A. D. Alhaidari, Solutions of the nonrelativistic wave equation with position-dependent effective mass. Phys. Rev. A 66, 042116 (2002).DOI
  3. [3]G. Junker Supersymmetric Methods in Quantum, Statistical and Solid State Physics. IOP Publishing (2019).DOI
  4. [4]J. Sesma, Exact solution of the Schrödinger equation with a Lennard-Jones potential. J Math Chem 51, 1881 - 1896 (2013). https://doi.org/10.1007/s10910-013-0189-9.
  5. [5]Daniel Baye, The Lagrange-mesh method. Physics Reports Vol. 565, 1 -107 (2015).DOI
  6. [6]A.D. Alhaidari, I.A. Assi, and A. Mebirouk. Bound states of a quartic and sextic inverse-power-law potential for all angular momenta. The European Physical Journal Plus, 136(4), 1-12 (2021).DOI
  7. [7]A. Mebirouk, S. Bouheroum-Mentri and L. Aceto, Approximation of eigenvalues of Sturm-Liouville problems defined on a semi-infinite domain, Appl. Math. Comput. 369 124823 (2020).DOI
  8. [8]H. Ciftci, R. L. Hall, and N. Saad, J. Phys. A: Math. & Gen., 36 (47), 11807 (2003).
  9. [9]B. Champion, R.L. Hall, and N. Saad. Asymptotic iteration method for singular potentials. International Journal of Modern Physics A, 23(09), 1405-1415 (2008).DOI
  10. [10]T. Barakat. The asymptotic iteration method for the eigenenergies of the Schrödinger equation with the potential V(r) = −Z/r + gr + λr². Journal of Physics A: Mathematical and General, 39(4), 823 (2006).DOI
  11. [11]P. Amore and F.M. Fernández, Comment on an application of the asymptotic iteration method to a perturbed Coulomb model. Journal of Physics A: Mathematical and General, 39(33), 10491 (2006).DOI
  12. [12]H.I. Alrebdi and T. Barakat. Optimal perturbation technique within the Asymptotic Iteration Method for heavy-light meson mass splittings. Universe, 7(6), 180 (2021).DOI
  13. [13]F. M. Fernandez. Introduction to Perturbation Theory in Quantum Mechanics (1st ed.). CRC Press (2000). https://doi.org/10.1201/9781420039641
  14. [14]R. L. Hall, N. Saad, and A. B. von Keviczky, Generalized spiked harmonic oscillator. J. Phys. A 34, 1169 (2001).DOI
  15. [15]R. L. Hall, N. Saad, and A. B. Keviczky, Spiked harmonic oscillators, J. Math. Phys. 43, 94 (2002).
  16. [16]F. J. Gómez and J. Sesma, Spiked oscillators: exact solution. J. Phys. A 43, 385302 (2010).
  17. [17]A. K. Roy, Ground and excited states of spherically symmetric potentials through an imaginary-time evolution method: application to spiked harmonic oscillators. J. Math. Chem. 52, 2645 (2014).arxiv.org/abs/1801.05647
  18. [18]A. K. Roy and A. F. Jalbout, Bound states of the generalized spiked harmonic oscillator, J. Molec. Struct. 853, 27 (2008).arxiv.org/abs/1307.3323
  19. [19]O. Mustafa and M. Odeh, Bound states for spiked harmonic oscillators and truncated Coulomb potentials, J. Phys. A 32, 3055 (1999).DOI
  20. [20]A.D. Alhaidari. Finite series representation for the bound states of a spiked isotropic oscillator with inverse-quartic singularity. Modern Physics Letters A, 37(07), 2250045 (2022).DOI

Paper details

vlnc:W_2vnx7rz9
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC0-1.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents