This paper was converted on www.awesomepapers.org from LaTeX by an anonymous user.
Want to know more? Visit the Converter page.

Normalized eigenfunctions of parametrically factored Schrödinger equations

J. de la Cruz and H.C. Rosu Instituto Potosino de Investigación Científica y Tecnológica, Camino a la Presa San José 2055, Col. Lomas 4a Sección, San Luis Potosí, 78216 S.L.P., Mexico [email protected]; ORCID: 0000-0001-5943-5752 [email protected]; ORCID: 0000-0001-5909-1945
Abstract

The factorizations using the general Riccati solution constructed from a given particular solution by means of the Bernoulli ansatz initiated in 1984 by Mielnik and Fernández C. for the cases of the quantum harmonic oscillator and the radial Hydrogen equation, respectively, are briefly reviewed. The issue of the eigenfunction normalization of the obtained one-parameter Darboux-deformed Hamiltonians is addressed here.

1 Introduction

According to Infeld and Hull [1], the factorization method for Sturm-Liouville differential equations owes its existence primarily to Schrödinger, who at the beginning of 1940’s discussed the method [2] and also factored in four different ways the hypergeometric differential equation [3]. However, they also mention Weyl [4] and Dirac [5] as precursors, who, a decade earlier, considered the spherical harmonics with spin and the creation and annihilation operators, respectively, in a similar framework. Dirac himself gave credit to Fock for the latter factoring operators. After the comprehensive paper of Infeld and Hull from 1951, there were only very few works, e.g., Crum [6], related to the factorization methods that have been published along the next three decades. However, during the 1980’s, with the advent of supersymmetric quantum mechanics, the factorizations in their supersymmetric version with the usage of particular Riccati solutions have become the preferred tool for generating isospectral potentials with different degrees of isospectrality. On the other hand, although the factorizations based on general Riccati solutions for the fundamental cases of quantum harmonic oscillator and the Hydrogen atom have been also addressed in the same period in the pioneering papers of Mielnik [7] and Fernández C. [8], respectively, they really did not enter the mainstream research of supersymmetric quantum mechanics, although one can see chapter 6 in [9] and references therein and in the review paper [10]. In fact, more attention attracted their connection with the version of the Gel’fand-Levitan inverse method developed by Abraham and Moses [11], as mentioned in [7, 8] and also noticed by Nieto [12]. In a parallel line of research, after Matveev [13, 14] revived the Darboux transformations [15], Andrianov et al. [16, 17] showed their equivalence with the supersymmetric factorization method and therefore one may also consider the parametric factorizations as a particular kind of parametric Darboux transformations.

The main goal of this short paper, which is dedicated to Professor D.J. Fernández on the occasion of his forty years of remarkable scientific activity, is to revisit the pioneering papers of 1984, pinpointing some details, such as the issue of normalization constants of the parametrically deformed wavefunctions, which has not been mentioned in the original papers.

The rest of the paper is structured as follows. In Section 2, the parametric factorization of a Schrödinger like equation is presented in general terms. Section 3 is devoted to the application to the quantum harmonic oscillator following [7], with a subsection on the normalization constant of the deformed oscillator eigenfunctions. The application to the radial equation of the Hydrogen atom following [8], with a similar subsection on the normalization constants, is presented in Section 4. The Conclusions section ends up the paper.

2 The one-parameter factorization

Let us consider the one-dimensional Schrödinger equation

(D2+f(x)E)F=0,D2=d2dx2,\left(-D^{2}+f(x)-E\right)F=0~{},\qquad D^{2}=\frac{d^{2}}{dx^{2}}~{}, (1)

where the Schrödinger potential f(x)f(x), is an arbitrary C2C^{2} function, and also primes will be used in the following for the derivatives of some functions. We set E=0E=0 for the ground state energy, thus turning the equation into a homogeneous differential equation for the ground state wavefunction F0F_{0}

(D2+f(x))F0=0.\left(-D^{2}+f(x)\right)F_{0}=0~{}. (2)

Equation (2) can be factored in the form

(D+Φ)(D+Φ)F0[D2+(Φ2Φ)]F0=0,(-D+\Phi)(D+\Phi)F_{0}\equiv\big{[}-D^{2}+(\Phi^{2}-\Phi^{\prime})\big{]}F_{0}=0~{},\qquad (3)

where Φ=F0/F0\Phi=-F_{0}^{\prime}/F_{0} is the negative logarithmic derivative of a solution F0F_{0} of (2), a.k.a. the seed solution of the Darboux transformation (DT). Comparing (2) and (3), one obtains the Riccati equation,

Φ2Φ=f(x).\Phi^{2}-\Phi^{\prime}=f(x)~{}. (4)

If one knows a Riccati solution Φ\Phi of (4), the solution of (2) can be obtained from F0=exp(xΦ)F_{0}={\rm exp}(-\int^{x}\Phi).

The non-parametric Darboux-transformed equation of (3), also known as the supersymmetric partner equation of (3), is obtained by reverting the factoring

(D+Φ)(D+Φ)F^0[D2+(Φ2+Φ)]F^0=[D2+(f+2Φ)]F^0=0.(D+\Phi)(-D+\Phi)\hat{F}_{0}\equiv\big{[}-D^{2}+(\Phi^{2}+\Phi^{\prime})\big{]}\hat{F}_{0}=\big{[}-D^{2}+(f+2\Phi^{\prime})\big{]}\hat{F}_{0}=0~{}. (5)

The generic interesting fact of the reverted factorizations is that they are not unique, but parametric. This is because in the factorization brackets one can use the general Riccati solution in the form of the Bernoulli ansatz Φg=Φ+1/u\Phi_{g}=\Phi+1/u, where the function uu satisfies the first-order differential equation

u+2Φu+1=0-u^{\prime}+2\Phi u+1=0 (6)

and not just a particular solution Φ\Phi.

Then, it is easy to show that

(D+Φ+1u)(D+Φ+1u)F¯0=[D2+(Φ2+Φ)]F¯0=0.\left(D+\Phi+\frac{1}{u}\right)\left(-D+\Phi+\frac{1}{u}\right)\bar{F}_{0}=\big{[}-D^{2}+(\Phi^{2}+\Phi^{\prime})\big{]}\bar{F}_{0}=0~{}. (7)

Furthermore, the left hand side of the latter equation can be written as

[D2+(Φg2+Φg)]F¯0=0.\big{[}-D^{2}+\left(\Phi^{2}_{g}+\Phi^{\prime}_{g}\right)\big{]}\bar{F}_{0}=0~{}. (8)

Therefore, F¯0=F^0\bar{F}_{0}=\hat{F}_{0}, and it does not matter if one uses the particular or general Riccati solution.

The relevant result is obtained only when one reverts back the factorization brackets in the intent to return to the initial equation. Then, one obtains

[D2+(Φg2Φg)]~0=0.\big{[}-D^{2}+\left(\Phi^{2}_{g}-\Phi^{\prime}_{g}\right)\big{]}\widetilde{{\cal F}}_{0}=0~{}. (9)

Substituting the Bernoulli form of Φg\Phi_{g} in (9) leads to

(D2+f(x)+4Φu1+2u2)~0=0.\left(-D^{2}+f(x)+4\Phi u^{-1}+2u^{-2}\right)\widetilde{{\cal F}}_{0}=0~{}. (10)

The latter equation is actually a one-parameter family of equations having the same Darboux-transformed partner, the running parameter of the family being the integration constant that occurs in Bernoulli’s function 1/u1/u obtained by the integration of (6)

1u=e2xΦ(x)𝑑xγ+xe2x¯Φ(x)𝑑x𝑑x¯=ddxln(γ+xe2xΦ(x)𝑑x𝑑x).\frac{1}{u}=\frac{e^{-2\int^{x}\Phi(x^{\prime})dx^{\prime}}}{\gamma+\int^{x}e^{-2\int^{\bar{x}}\Phi(x^{\prime})dx^{\prime}}d\bar{x}}=\frac{d}{dx}\ln\left(\gamma+\int^{x}e^{-2\int^{x}\Phi(x^{\prime})dx^{\prime}}dx\right)~{}. (11)

It is easy to show that equation (10) differs from the initial equation (1) by

4Φ/u+2/u2=21uddxln(1u),4\Phi/u+2/u^{2}=2\frac{1}{u}\frac{d}{dx}\ln\left(\frac{1}{u}\right)~{}, (12)

which is the additive Darboux deformation of f(x)f(x) introduced by the parametric DT. Moreover, the solutions ~0\widetilde{{\cal F}}_{0} of the parametric Darboux-transformed equations are related to the undeformed solutions F0F_{0} as follows

~0=exΦg𝑑s=exΦ𝑑sexddxln(γ+xe2xΦ(s)𝑑s)=F0γ+xF02(s)𝑑s.\widetilde{{\cal F}}_{0}=e^{-\int^{x}\Phi_{g}ds}=e^{-\int^{x}\Phi ds}e^{-\int^{x}\frac{d}{dx}\ln\left(\gamma+\int^{x}e^{-2\int^{x}\Phi(s)ds}\right)}=\frac{F_{0}}{\gamma+\int^{x}F_{0}^{2}(s)ds}~{}. (13)

Before ending this section, we mention that including a scalar scaling cc in front of D2D^{2} can be manipulated by democratically distributing c\sqrt{c} in front of each of the factoring brackets (±D+Φ)(\pm D+\Phi). The only change is the scaling by cc of the eigenspectrum of the given Schrödinger problem. Therefore, in the following we will use the usual quantum-mechanical scaling 1/21/\sqrt{2} in front of the factoring operators.

3 The Quantum Harmonic Oscillator (Φ=x\Phi=x)

For this case, the factorization brackets are similar to Fock’s creation and annihilation operators a=12(ddx+x)a=\frac{1}{\sqrt{2}}\left(\frac{d~{}}{dx}+x\right) and its adjoint conjugate aa^{\dagger}, which factorize the Hamiltonian Hqho=12d2dx2+12x2H_{qho}=-\frac{1}{2}\frac{d^{2}~{}}{dx^{2}}+\frac{1}{2}x^{2} as

aa=H+12,andaa=H12,aa^{\dagger}=H+\frac{1}{2}~{},\quad{\rm and}\quad a^{\dagger}a=H-\frac{1}{2}~{}, (14)

are replaced by the new operators b=12(ddx+β(x))b=\frac{1}{\sqrt{2}}\left(\frac{d~{}}{dx}+\beta(x)\right), and its adjoint conjugate bb^{\dagger}, requiring that

bb=H+12,bb=H~12.bb^{\dagger}=H+\frac{1}{2}~{},\quad b^{\dagger}b=\widetilde{H}-\frac{1}{2}~{}. (15)

In order to satisfy this, β(x)\beta(x) is found to satisfy the Riccati equation

β+β2=1+x2\beta^{\prime}+\beta^{2}=1+x^{2} (16)

and the reversed factorization bbb^{\dagger}b provides the one-parameter isospectral Hamiltonian with the one-parameter contribution to the potential

H~=H+ddx[ex2γ+xex2𝑑x],\widetilde{H}=H+\frac{d~{}}{dx}\left[\frac{e^{-x^{2}}}{\gamma+\int_{-\infty}^{x}e^{-x^{\prime 2}}dx^{\prime}}\right]~{}, (17)

where the real parameter γ\gamma is required to be |γ|>12π|\gamma|>\frac{1}{2}\sqrt{\pi} if we want to avoid singular potentials.

From (17) and (13) with F0=ex22F_{0}=e^{-\frac{x^{2}}{2}} and ~0=Ψ~0(x;γ)\widetilde{{\cal F}}_{0}=\widetilde{\Psi}_{0}(x;\gamma), one obtains

V~(x;γ)=x22+ddx(ex2γ+xex2𝑑x),Ψ~0(x;γ)=ex2/2γ+xex2𝑑x,\widetilde{V}(x;\gamma)=\frac{x^{2}}{2}+\frac{d}{dx}\left(\frac{e^{-x^{2}}}{\gamma+\int_{-\infty}^{x}e^{-x^{\prime 2}}dx^{\prime}}\right)~{},\quad\widetilde{\Psi}_{0}(x;\gamma)=\frac{e^{-x^{2}/2}}{\gamma+\int_{-\infty}^{x}e^{-x^{\prime 2}}dx^{\prime}}~{}, (18)

respectively. In Fig. 2, one can observe plots of singular and regular one-parameter deformed counterparts of the harmonic oscillator potential and ground state for γ=0.5\gamma=\mp 0.5, respectively.

Refer to caption
Refer to caption
Figure 1: Singular (red) and regular (blue) Mielnik potentials and ground state wavefunctions Ψ~0(x;γ)\widetilde{\Psi}_{0}(x;\gamma) for γ=0.5\gamma=\mp 0.5. The undeformed harmonic oscillator potential and its unnormalized ground state Gaussian wavefunction are also included in black color for comparison as it is done with the appropriate undeformed cases in all the plots of the paper.

3.1 The normalization constant of the parametrically-deformed oscillator wavefunctions

To determine the normalization constant of the one-parameter Darboux-deformed oscillator ground state, we write the normalization integral

N02ex2dx(γ+xex2𝑑x)2=1,\int_{-\infty}^{\infty}N_{0}^{2}\frac{e^{-x^{\prime 2}}dx^{\prime}}{(\gamma+\int_{-\infty}^{x}e^{-x^{\prime 2}}dx^{\prime})^{2}}=1~{}, (19)

which, by introducing the change of variable X(x)=xex2𝑑xX(x)=\int_{-\infty}^{x}e^{-x^{\prime 2}}dx^{\prime}, becomes

0πN02dX(γ+X)2=1.\int_{0}^{\sqrt{\pi}}N_{0}^{2}\frac{dX}{(\gamma+X)^{2}}=1~{}. (20)

Effecting the integral gives

N0N0(γ)=γ(γπ+1).N_{0}\equiv N_{0}(\gamma)=\sqrt{\gamma\left(\frac{\gamma}{\sqrt{\pi}}+1\right)}~{}. (21)

This shows that γ\gamma should not be in [π,0][-\sqrt{\pi},0]. In Fig. 2, we present plots of two regular Darboux-parametric potentials and the corresponding normalized eigenfunctions.

Refer to caption
Refer to caption
Figure 2: Regular Mielnik potentials and corresponding normalized ground state wavefunctions ψ~0(x;γ)=N0(γ)Ψ~0(x;γ)\tilde{\psi}_{0}(x;\gamma)=N_{0}(\gamma)\widetilde{\Psi}_{0}(x;\gamma) for γ=0.5\gamma=0.5 and 11, red and blue, respectively.

Another interesting fact about the normalization of the Darboux-deformed ground states is that there is a pair of parameters for which the normalization constant is exactly that of the original ground state, despite the deformed wavefunctions not being Gaussians. In the case of the harmonic oscillator, this pair is given by

γ±=12(π±π+4)\gamma_{\pm}=\frac{1}{2}\left(-\sqrt{\pi}\pm\sqrt{\pi+4}\right) (22)

for which N0(γ±)=π1/4N_{0}(\gamma_{\pm})=\pi^{-1/4}. These cases are presented in Fig. 3, where one can notice the symmetries V~(x;γ)=V~(x;γ+)\tilde{V}(x;\gamma_{-})=\tilde{V}(-x;\gamma_{+}) and ψ~(x;γ)=ψ~(x;γ+)\tilde{\psi}(x;\gamma_{-})=-\tilde{\psi}(-x;\gamma_{+}). For other values of the parameter, the normalization constant can be very big. For example, for γm=mπ\gamma_{m}=m\sqrt{\pi}, one obtains N0(m)=m(m+1)π1/4N_{0}(m)=\sqrt{m(m+1)}\pi^{1/4}.

Refer to caption
Refer to caption
Figure 3: The one-parameter deformed oscillator potentials and corresponding normalized ground state wavefunctions for γ±\gamma_{\pm} from (22).

4 The Radial Hydrogen Equation (Φ=r1\Phi=\frac{\ell}{r}-\frac{1}{\ell})

The radial Schrödinger equation for the Hydrogen atom may be written as the following Sturm-Liouville eigenvalue problem

HR(r)=1r[d2dr2+(+1)r22r]rR(r)=λnR(r),H_{\ell}\,R(r)=\frac{1}{r}\left[-\frac{d^{2}~{}}{dr^{2}}+\frac{\ell(\ell+1)}{r^{2}}-\frac{2}{r}\right]r\,R(r)=\lambda_{n}R(r)~{}, (23)

where HH_{\ell} is the radial Hamiltonian operator and the used units are m==e=4πε0=1m=\hbar=e=4\pi\varepsilon_{0}=1, together with ZZ=1.

With this convention, the energy eigenvalues and the eigenfunctions can be written in the form

λn=1n2,Rn,(r)=Cn,(2rn)ernLn12+1(2rn),\lambda_{n}=-\frac{1}{n^{2}}\,,\ \ \ R_{n,\ell}(r)=C_{n,\ell}\ \left(\frac{2r}{n}\right)^{\ell}\,e^{-\frac{r}{n}}L^{2\ell+1}_{n-\ell-1}\left(\frac{2r}{n}\right)~{}, (24)

where Cn,C_{n,\ell} are the normalization constants given by

Cn,=2n2(n1)!(n+)!,C_{n,\ell}=\frac{2}{n^{2}}\sqrt{\frac{(n-\ell-1)!}{(n+\ell)!}}~{}, (25)

nn\in\mathbb{N}, n>n>\ell, and Ln12+1(2r/n)L^{2\ell+1}_{n-\ell-1}(2r/n) are derivatives of order 2+12\ell+1 of the Laguerre polynomials of degree n1n-\ell-1, a.k.a. associated Laguerre polynomials. The polynomial degree nr=n1n_{r}=n-\ell-1 provides the number of nodes of the eigenfunctions in the radial direction. In particular, we are interested in ’ground-state’ wavefunctions for which n=+1n=\ell+1.

The ladder operators of Infeld and Hull provide the following factorizations of the radial H operator (23)

a+a=\displaystyle a_{\ell}^{+}a_{\ell}^{-}= H+12,\displaystyle H_{\ell}+\frac{1}{\ell^{2}}~{}, (26)
a+1a+1+=\displaystyle a_{\ell+1}^{-}a_{\ell+1}^{+}= H+1(+1)2,\displaystyle H_{\ell}+\frac{1}{({\ell+1})^{2}}\ , (27)

where the Infeld-Hull ladder operators act on the radial functions as follows:

aRn,\displaystyle a^{-}_{\ell}R_{n,\mathop{\ell}\nolimits} =1r(ddr+r1)rRn,=cnRn,1,\displaystyle=\frac{1}{r}\left(\frac{d}{dr}+\frac{\ell}{r}-\frac{1}{\ell}\right)r\,R_{n,\ell}=c_{n\ell}\,R_{n,\ell-1}~{}, (28)
a+Rn,1\displaystyle a^{+}_{\ell}R_{n,\ell-1} =1r(ddr+r1)rRn,1=cnRn,,\displaystyle=\frac{1}{r}\left(-\frac{d}{dr}+\frac{\ell}{r}-\frac{1}{\ell}\right)rR_{n,\ell-1}=c_{n\ell}\,R_{n,\mathop{\ell}\nolimits}~{}, (29)

where cn=(n)(n+)nc_{n\ell}=\frac{\sqrt{(n-\ell)(n+\ell)}}{n\mathop{\ell}\nolimits}.


Using Mielnik’s factorization approach, Fernández C. introduced the following modified Infeld-Hull raising and lowering operators [8]

A+=1r(ddr+β)r,A=1r(ddr+β)r,A+A=H+12,AA+=H~1+12,\begin{split}&A^{+}_{\ell}=\frac{1}{r}\left(-\frac{d}{dr}+\beta_{\ell}\right)r~{},\quad A^{-}_{\ell}=\frac{1}{r}\left(\frac{d}{dr}+\beta_{\ell}\right)r~{},\\ &A^{+}_{\ell}A^{-}_{\ell}=H_{\ell}+\frac{1}{\ell^{2}}~{},\qquad A^{-}_{\ell}A^{+}_{\ell}=\widetilde{H}_{\ell-1}+\frac{1}{{\ell}^{2}}~{},\end{split} (30)

where

β=r1+r2e2r/γ0ry2e2y/𝑑y,\beta_{\ell}=\frac{\ell}{r}-\frac{1}{\ell}+\frac{r^{2\ell}e^{-2r/\ell}}{\gamma_{\ell}-\int_{0}^{r}y^{2\ell}e^{-2y/\ell}dy}~{}, (31)

obtained as the general solution of the Riccati equation

β+β2=2r+(+1)r2+1l2.-\beta^{\prime}_{\ell}+\beta_{\ell}^{2}=-\frac{2}{r}+\frac{\mathop{\ell}\nolimits(\mathop{\ell}\nolimits+1)}{r^{2}}+\frac{1}{l^{2}}~{}. (32)

The parametric radial Hydrogen potentials and ground state wavefunctions (for the eigenvalue 1/2-1/\ell^{2}) are given by

V~1(r;γ)=2r+(1)r2+ddr(2r2e2r/γ0ry2e2y/𝑑y),\widetilde{V}_{\ell-1}(r;\gamma)=-\frac{2}{r}+\frac{\mathop{\ell}\nolimits(\mathop{\ell}\nolimits-1)}{r^{2}}+\frac{d}{dr}\left(\frac{2r^{2\ell}e^{-2r/\ell}}{\gamma_{\ell}-\int_{0}^{r}y^{2\ell}e^{-2y/\ell}dy}\right)~{}, (33)
R~,1=r1er/γ0rr2e2r/𝑑r,\widetilde{R}_{\ell,\ell-1}=\frac{r^{\ell-1}e^{-r/\ell}}{\gamma_{\ell}-\int_{0}^{r}r^{\prime 2\ell}e^{-2r^{\prime}/\ell}dr^{\prime}}~{}, (34)

where

γ>(2)!(2)2+1,orγ<0,\gamma_{\ell}>(2\ell)!\left(\frac{\ell}{2}\right)^{2\ell+1}~{},\ \ \mbox{or}\ \ \gamma_{\ell}<0~{}, (35)

to avoid singularities.


4.1 Normalization constants for R~,1\widetilde{R}_{\ell,\ell-1}

In this case, the normalization integral is

0Nl2R~,12r2𝑑r0Nl2r2e2r/(γ0rr2e2r/𝑑r)2𝑑r=1.\int_{0}^{\infty}N_{l}^{2}\widetilde{R}_{\ell,\ell-1}^{2}r^{2}dr\equiv\int_{0}^{\infty}N_{l}^{2}\frac{r^{2\ell}e^{-2r/\ell}}{\left(\gamma_{\ell}-\int_{0}^{r}r^{\prime 2\ell}e^{-2r^{\prime}/\ell}dr^{\prime}\right)^{2}}dr=1~{}. (36)

The change of variable similar to the case of harmonic oscillator

Γ(r)=0rr2e2r/𝑑r\Gamma_{\ell}(r)=\int_{0}^{r}r^{\prime 2\ell}e^{-2r^{\prime}/\ell}dr^{\prime} (37)

requires the calculation of Γ()=Γ\Gamma_{\ell}(\infty)=\Gamma_{\ell}, which can be performed through recursive integrations by parts providing the result:

Γ=(2)!(2)2+1.\Gamma_{\ell}=(2\ell)!\left(\frac{\ell}{2}\right)^{2\ell+1}~{}. (38)

The normalization constant is then given by

N=γ(γΓ1),N_{\ell}=\sqrt{\gamma_{\ell}\left(\frac{\gamma_{\ell}}{\Gamma_{\ell}}-1\right)}~{}, (39)

and γ\gamma_{\ell} can take any arbitrary negative values, while if taken positive, should be strictly greater than Γ\Gamma_{\ell}, confirming (35).

Thus:

For =1\ell=1, i.e., for R~1,0\widetilde{R}_{1,0}, we have N1=γ1(4γ11)N_{1}=\sqrt{\gamma_{1}\left(4\gamma_{1}-1\right)}, and one should take γ1>1/4\gamma_{1}>1/4.

For =2\ell=2, i.e., for R~2,1\widetilde{R}_{2,1}, we have N2=γ2(γ2241)N_{2}=\sqrt{\gamma_{2}\left(\frac{\gamma_{2}}{24}-1\right)}, and one should take γ2>24\gamma_{2}>24, and so forth.

The one-parameter potentials and the corresponding normalized eigenfunctions are presented in Fig. 4 for =1,2,3\ell=1,2,3 together with the undeformed cases for comparison. At high values of γ\gamma_{\ell} the supplementary pockets of the deformed potentials tend towards the nondeformed Coulombian well with centrifugal barrier, or without it for the case =1\ell=1, corresponding to the eigenvalue 1/2-1/\ell^{2} as expected.

If we ask the deformed wavefunctions to have the same normalization as the undeformed wavefunctions, one can obtain from (25) that

C,1C=2l2(2l1)!,C_{\ell,\ell-1}\equiv C_{\ell}=\frac{2}{l^{2}\sqrt{(2l-1)!}}~{},

which then provides a pair of deformation parameters given by the roots of the equation

γ2ΓγΓC2=0.\gamma^{2}-\Gamma_{\ell}\gamma-\Gamma_{\ell}C_{\ell}^{2}=0~{}. (40)

For example, from the latter equation, one obtains

γ±=18(1±65)andγ±=26(6±6+124)\gamma_{\pm}=\frac{1}{8}\left(1\pm\sqrt{65}\right)\qquad{\rm and}\qquad\gamma_{\pm}=2\sqrt{6}\left(\sqrt{6}\pm\sqrt{6+\frac{1}{24}}\right) (41)

for =1\ell=1 and =2\ell=2, respectively. Plots of the deformed potentials and normalized wavefunctions for these two cases are presented in Fig. 5.


5 Conclusions

We have revisited the pioneering papers of Mielnik and Fernández that opened the research of parametric families of isospectral potentials in supersymmetric quantum mechanics. The uncovered issue of normalization of the parametric deformed ground state wavefunctions is discussed in some detail. It is found that there exist pairs of deformation parameters (one positive, the other negative) for which the deformed wavefunctions have the same normalization constant as the undeformed wavefunctions.


Acknowledgment

The second author wishes to thank the organizers of the workshop honoring Professor David J. Fernández C. for the enjoyable days of the event. Thanks also to Professors David Fernández and Asim Gangopadhyaya for some clarifying remarks.

Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 4: Parametric Hydrogen radial potentials and corresponding normalized one-parameter R~,1(r;γ)\tilde{R}_{\ell,\ell-1}(r;\gamma) eigenfunctions for =1,2,3\ell=1,2,3.
Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 5: The pairs of Hydrogen deformed radial potentials V~1\tilde{V}_{\ell-1} and corresponding deformed wavefunctions R~,1(r;γ)\tilde{R}_{\ell,\ell-1}(r;\gamma) having the same normalization constant as the undeformed wavefunctions, for =1,2\ell=1,2.

References

  • [1] Infeld L. and Hull T.E. 1951 The factorization method, Rev. Mod. Phys. 23, 21–68. 10.1103/RevModPhys.23.21
  • [2] Schrödinger E. 1940 A method of determining quantum-mechanical eigenvalues and eigenfunctions, Proc. Roy. Irish Acad. A 46, 9-16 and 183-206.
  • [3] Schrödinger E. 1941 The factorization of the hypergeometric equation, Proc. Roy. Irish Acad. A 46, 53-54. 10.48550/arXiv.physics/9910003
  • [4] Weyl H. 1931 The Theory of Groups and Quantum Mechanics 2nd ed (English) (E. P. Dutton and Company, Inc., New York)
  • [5] Dirac P.A.M. 1930 The Principles of Quantum Mechanics (Oxford Clarendon Press, UK)
  • [6] Crum M.M. 1955 Associated Sturm-Liouville systems, Quart. J. Math. Oxford 6 121-127. 10.1093/qmath/6.1.121
  • [7] Mielnik B. 1984 Factorization method and new potentials with oscillator spectrum, J. Math. Phys. 25, 3387-3389. 10.1063/1.526108
  • [8] Fernández D.J. 1984 New Hydrogen-like potentials, Lett. Math. Phys. 8, 337–343. 10.1007/BF00400506 ; arXiv:quant-ph/0006119. 10.48550/arXiv.quant-ph/0006119
  • [9] Cooper F. Khare A. and Sukhatme U. 2001 Supersymmetry in Quantum Mechanics (World Scientific, Singapore)
  • [10] Mielnik B. and Rosas-Ortiz O. 2004 Factorization: little or great algorithm ?, J. Phys. A: Math. Gen. 37, 10007–10035. 10.1088/0305-4470/37/43/001
  • [11] Abraham P.B. and Moses H.E. 1980 Changes in potentials due to changes in the point spectrum: Anharmonic oscillators with exact solutions, Phys. Rev. A 22, 1333–1340. 10.1103/PhysRevA.22.1333
  • [12] Nieto M.M. 1984 Relationship between supersymmetry and the inverse method in quantum mechanics, Phys. Lett. B 145, 208–210. 10.1016/0370-2693(84)90339-3
  • [13] Matveev V.B. 1979 Darboux transformation and the explicit solutions of differential-difference and difference-difference evolution equations I, Lett. Math. Phys. 3, 217-22. 10.1007/BF00405296
  • [14] Matveev V.B. and Salle M.A. 1979 Darboux transformation and the explicit solutions of differential-difference and difference-difference evolution equations II, Lett. Math. Phys. 3, 425-429. 10.1007/BF00397217
  • [15] Darboux G. 1882 Sur une proposition relative aux équations linéaires, C. R. Acad. Sci. Paris 94, 1456-1459. 10.48550/arXiv.physics/9908003
  • [16] Andrianov A.A. Borisov N.V. and Ioffe M.V. 1984 The factorization method and quantum systems with equivalent energy spectra, Phys. Lett. A 105, 19-22. 10.1016/0375-9601(84)90553-X
  • [17] Andrianov A.A. Borisov N.V. and Ioffe M.V. 1984 Factorization method and Darboux transformation for multidimensional Hamiltonians, Theor. Math. Phys. 61, 1078-1088. 10.1007/BF01029109