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

Coherent states on a circle: the Higgs-like approach

Ali Mahdifar1,2 [email protected] Ehsan Amooghorban3,4 [email protected] 1 Department of Physics, University of Isfahan, Hezar Jerib, Isfahan, 81746-73441, Iran. 2 Department of Physics, Quantum optics group, University of Isfahan, Hezar Jerib St. Isfahan 81764-73441, Iran. 3 Department of Physics, Faculty of Science, Shahrekord University, Shahrekord, 88186-34141, Iran. 4 Nanotechnology Research Center, Shahrekord University, Shahrekord 88186-34141, Iran.
Abstract

In this paper, the Higgs-like approach is used to analyze the quantum dynamics of a harmonic oscillator constrained on a circle. We obtain the Hamiltonian of this system as a function of the Cartesian coordinate of the tangent line through the gnomonic projection and then quantize it in the standard way. We then recast the Hamiltonian in a shape-invariant form and derive the spectrum energy of the confined harmonic oscillator on the circle. With help of the f-deformed oscillator algebra, we construct the coherent states on the circle and investigate their quantum statistical properties. We find that such states show nonclassical features like squeezing and sub-Poissonian statistics even in small curvatures of the circle.

keywords:
Quantum Harmonic Oscillator on a Circle, Shape-Invariant Hamiltonian, Coherent States, Quantum Statistical Properties.
PACS:
03.65.Fd, 42.50.Dv

1 Introduction

One of the fundamental problems in quantum mechanics is to find a compatible quantization procedure of a classical system. The canonical quantization is the usual well-known approach which is based on using the classical form of observables as a real function in phase space and replacing the coordinates xx and pp with operators q^\hat{q} and p^\hat{p} in a symmetrical order. However, this procedure works only for systems with phase space R2NR^{2N}, provided that the quantization is carried out in Cartesian coordinates. In a non-Cartesian coordinate, we can start from the Lagrangian formalism and obtain canonical conjugate momenta of the corresponding dynamical variables of the system to construct the Hamiltonian, and impose canonical commutation relations between them. However, it raises a host of difficulties like a mismatch between the canonical momenta and the Noether momenta [1]. Instead, we can consistently quantize the system in Cartesian coordinates and make the change to curvilinear coordinates at the quantum level. In general, the canonical quantization becomes cumbersome when the classical coordinates and/or their conjugate momenta do not vary over the full interval from -\infty to ++\infty.

There are various reasons why physicists try to develop the correct quantum mechanics in curved spaces. Despite extending our knowledge on quantum mechanics in more general mathematical frameworks, it provides the capacity for application in many different fields of physics, such as the quantum Hall effect [2], the quantum dots [3, 4], and the coherent state quantization [5]. To the best of our knowledge, there are three known approaches to study nonrelativistic quantum mechanics in the spaces of constant curvature: The first one is the Noether quantization which is concerned with the identification of the Killing vector fields and the Noether momenta. In this manner, one can construct the Hamiltonian of the system as a function of Noether momenta, and then apply the quantization process to the components of these momenta [6, 7, 8]. The second method is the so-called thin layer quantization [9, 10, 11, 12]. In this approach, the 2D curved surface is embedded into the larger 3D Euclidean space, and the introduction of effective potential results in a dimensional reduction in the Schrodinger equation. In the last approach, which is known as the Higgs method, dynamical symmetries of a system with a spherical geometry are extracted[13, 14]. In this way, the Hamiltonian of the system is expressed in terms of the Casimir operators of the Lie algebra of such dynamical symmetries, and subsequently, the energy eigenvalues are obtained through the eigenvalues of the Casimir operators. In this contribution, we follow the Higgs-like approach.

On the other hand, the standard coherent states (CSs) [15] as well as the generalized CSs associated with various algebras [16, 17, 18] play an important role in different fields of physics. Among them, the so-called nonlinear or f-deformed CSs [19, 20] have attracted much interest due to their nonclassical properties, such as quantum interference and amplitude squeezing [21, 22]. Such states indeed are the eigenstates of the annihilation operator of the f-deformed oscillator [19, 20] can be realized in the center of mass motion of a laser-driven trapped ion [22, 23], and in a micromaser under intensity-dependent atom-field interaction [24]. It is well known, the resolution of identity is the most important mathematical feature of the CSs, i.e., they form an overcomplete set. Furthermore, the CSs include properties of the space that the relevant system defined on it. Therefore, if we consider, for example, the harmonic oscillator on a curved space and construct the associated CSs, they contain the curvature effects. Thus, employing the curvature-dependent CSs manifold makes it possible to investigate the curvature effects on the physical phenomena within the framework of the CSs approach [25]. This can enable us to study the curvature effects of some physical structures like nanostructures on their electronic and optical properties [26].

In Refs. [27] and [28], based on the Higgs model, the generalized (nonlinear) coherent states of a two-dimensional harmonic oscillator and a Kepler-Coulomb potential on a spherical surface are constructed and some of their quantum optical properties are studied. Later on, the thermal nonlinear coherent states on a sphere is presented in [29] and the relation of the nonclassical properties of the constructed CSs with the curvature and temperature is investigated.

We have recently presented the classical and quantum-mechanical treatment of the Higgs model in the presence of dissipation using a continuum of oscillators as a reservoir [30]. There, we showed that the transition probability between energy levels is dependent on the curvature of the physical space and the dissipative effect, and significant probabilities of transition are only possible if the transition and reservoir’s oscillators frequencies are almost in resonance. We have then investigated the dynamics of both a free particle and an isotropic harmonic oscillator constrained to move on a spheroidal surface using two consecutive gnomonic projections [31]. We have obtained the Hamiltonian of the systems as a function of the Cartesian coordinates of the tangent plane and finally quantized them. We have shown that the effect of nonsphericity of the surface appears like an effective potential. We found that the deviation from the sphericity leads to the split of the energy levels of the constrained oscillator on a sphere and lifts the degeneracy.

With the above background, the goal of this contribution is twofold. In the first step, we intend to obtain the classical and quantum Hamiltonian of a harmonic oscillator confined on a circle which is viewed as the simplest curvature in geometry. To do this, we use the gnomonic projection, which is a projection onto the tangent line from the center of the circle into the embedding line. We first obtain the metric of the circle background in the gnomonic coordinates and construct the Hamiltonian of the harmonic oscillator. We follow the canonical quantization method and derive the quantum counterpart of the classical Hamiltonian. We then recast the Hamiltonian in a shape-invariant form and compute the exact spectrum energy of the confined harmonic oscillator on a circle. We show that in the limit of the large radius of the circle (straight line), it properly reduces to the well-known energy spectrum of the conventional quantum harmonic oscillator. In the second step, we obtain a deformation function for such oscillator utilizing the f-deformed oscillator algebra. In the end, we construct the CSs on a circle employing the nonlinear CSs approach and investigate their quantum statistical properties.

This paper is organized as follows: In Sec. 2, by making use of the gnomonic projection onto the tangent line, we obtain the Hamiltonian of a harmonic oscillator confined on a circle. We subsequently quantize the aforementioned Hamiltonian using a factorization method and calculate its exact spectrum energy. In Sec. 3, we obtain a deformation function for the oscillator on a circle. In Sec. 4, we construct the corresponding CSs and examine their resolution of identity. The Sec.  5 is devoted to the study of the quantum statistical properties of the constructed CSs. Especially, the influence of curvature of the circle on their nonclassical properties is analyzed. Finally, the summary and concluding remarks are given in section 6.

2 Quantum harmonic oscillator on circle

In this section, we obtain the Hamiltonian of a harmonic oscillator confined on a circle. To this end, we obtain the metric of the circle background in the gnomonic coordinates and then construct the Hamiltonian of the harmonic oscillator.

Let us designate the Cartesian coordinates of the circle background by (q1,q2)(q_{1},q_{2}), and assume that they satisfy the circle equation,

q12+q22=R2,q_{1}^{2}+q_{2}^{2}=R^{2}, (1)

where, RR is the radius of the circle.

Refer to caption
Figure 1: Coordinate systems and projection from a circle with radius RR onto a line. The gnomonic projection of the point from the circle’s center onto the point xx on the tangent line.

If we denote the Cartesian coordinate of the tangent line to the circle by xx (see Fig. 1), the relationship between these two coordinates is given by

q1\displaystyle q_{1} =\displaystyle= x[1+λx2],\displaystyle\frac{x}{\sqrt{[1+\lambda x^{2}}]},
q2\displaystyle q_{2} =\displaystyle= 1λ[1+λx2],\displaystyle\frac{1}{\sqrt{\lambda[1+\lambda x^{2}}]}, (2)

where, λ=1R2\lambda=\frac{1}{R^{2}} is the curvature of the circle. Accordingly, a point on the circle can be represented as,

r(q1,q2)=(xΛ,1λΛ),\vec{r}\equiv(q_{1},q_{2})=(\frac{x}{\Lambda},\frac{1}{\sqrt{\lambda}\Lambda}), (3)

where,

Λ=1+λx2.\Lambda=\sqrt{1+\lambda x^{2}}. (4)

Now the differential of r\vec{r} is given by,

dr=rxdx=(1Λ3,λxΛ3)dx.d\vec{r}=\vec{r}_{x}dx=(\frac{1}{\Lambda^{3}},-\frac{\sqrt{\lambda}x}{\Lambda^{3}})dx. (5)

Here, rx\vec{r}_{x} is the derivatives of r\vec{r} with respect to xx. After a straightforward calculation, we obtain the metric of the circle as,

ds2\displaystyle ds^{2} \displaystyle\doteq drdr=(1Λ6+λx2Λ6)dx2=1Λ4dx2.\displaystyle d\vec{r}\cdot d\vec{r}=(\frac{1}{\Lambda^{6}}+\frac{\lambda x^{2}}{\Lambda^{6}})dx^{2}=\frac{1}{\Lambda^{4}}dx^{2}. (6)

From equation (6), we get

s˙2=x˙2Λ4.\displaystyle\dot{s}^{2}=\frac{\dot{x}^{2}}{\Lambda^{4}}. (7)

We can therefore write the Lagrangian of a harmonic oscillator on circular background as,

L\displaystyle L \displaystyle\doteq 12s˙2VHO(x)=12x˙2Λ4VHO(x),\displaystyle\frac{1}{2}\dot{s}^{2}-V_{\rm HO}(x)=\frac{1}{2}\frac{\dot{x}^{2}}{\Lambda^{4}}-V_{\rm HO}(x), (8)

where VHO(x)V_{\rm HO}(x) is the potential of the confined oscillator on the circle in terms of the tangent coordinate xx. In what follows, we obtain the aforementioned potential. Let ss be the arc-length of the geodesics from the north pole of the circle, r0=(0,R){\vec{r}}_{0}=(0,R), to the point r=(q1,q2){\vec{r}}=(q_{1},q_{2}). Using Eq.(1), we have

s2=q12+q22=gq12.s^{2}=q_{1}^{2}+q_{2}^{2}=g\ q_{1}^{2}. (9)

Here, the one-dimensional metric tensor on the circle is given by

g=1+q12R2q12.\displaystyle g=1+\frac{q_{1}^{2}}{R^{2}-q_{1}^{2}}. (10)

Accordingly, the potential energy for a harmonic oscillator on the circular curve can be written as (in this paper we put =m=ω=1)\hbar=m=\omega=1):

VHO(s)12s2=12gq12=(1+q12R2q12)q12.V_{\rm HO}(s)\equiv\frac{1}{2}s^{2}=\frac{1}{2}\ g\ q_{1}^{2}=(1+\frac{q_{1}^{2}}{R^{2}-q_{1}^{2}})q_{1}^{2}. (11)

Let us reconsider the gnomonic projection, which is the projection onto the tangent line from the center of the circle. By making use of Eq. (2), the points on the circle are expressed in terms of the coordinates of this projection. Thus, the potential of the oscillator on the circle in terms of xx is given by:

VHO(x)=12(1+λx2)(x21+λx2)=12x2.V_{\rm HO}(x)=\frac{1}{2}(1+\lambda x^{2})(\frac{x^{2}}{1+\lambda x^{2}})=\frac{1}{2}x^{2}. (12)

Now, by using the Lagrangian (8) and the potential (12), we can calculate the momentum as,

p=Lx˙=x˙Λ4.\displaystyle p=\frac{\partial L}{\partial\dot{x}}=\frac{\dot{x}}{\Lambda^{4}}. (13)

With proper calculation, we get the classical Hamiltonian of the harmonic oscillator constrained to the circular background as,

Hclx˙pL=12[(1+λx2)p]2+12x212π2+12x2,H_{\rm cl}\doteq\dot{x}p-L=\frac{1}{2}\left[(1+\lambda x^{2})p\right]^{2}+\frac{1}{2}x^{2}\equiv\frac{1}{2}\pi^{2}+\frac{1}{2}x^{2}, (14)

where π=(1+λx2)p\pi=(1+\lambda x^{2})p. By replacing the classical momentum pp by related operator (iddx)(-i\frac{d}{dx}), we can obtain the quantum counterpart of the classical Hamiltonian (14) as,

H\displaystyle H =\displaystyle= 12[(1+λx2)(iddx)(1+λx2)(iddx)]+12x2\displaystyle\frac{1}{2}\left[(1+\lambda x^{2})(-i\frac{d}{dx})(1+\lambda x^{2})(-i\frac{d}{dx})\right]+\frac{1}{2}x^{2} (15)
=\displaystyle= 12[(1+λx2)2d2dx22λx(1+λx2)ddx]+12x2.\displaystyle\frac{1}{2}\left[-(1+\lambda x^{2})^{2}\frac{d^{2}}{dx^{2}}-2\lambda x(1+\lambda x^{2})\frac{d}{dx}\right]+\frac{1}{2}x^{2}.

Now, for any real number γ\gamma we define the following linear operator:

A(γ)=12[ıπ+γx]=12[(1+λx2)ddx+γx].A(\gamma)=\frac{1}{\sqrt{2}}[\imath\pi+\gamma x]=\frac{1}{\sqrt{2}}\left[(1+\lambda x^{2})\frac{d}{dx}+\gamma x\right]. (16)

Indeed, the adjoint of A(γ)A(\gamma) is given by:

A(γ)=12[ıπ+γx]=12[(1+λx2)ddx+γx].A^{{\dagger}}(\gamma)=\frac{1}{\sqrt{2}}[-\imath\pi+\gamma x]=\frac{1}{\sqrt{2}}\left[-(1+\lambda x^{2})\frac{d}{dx}+\gamma x\right]. (17)

We can easily show that

H~1(γ)\displaystyle\tilde{H}_{1}(\gamma) :=\displaystyle:= A(γ)A(γ)\displaystyle A^{{\dagger}}(\gamma)A(\gamma)
=\displaystyle= 12[(1+λx2)2d2dx22λx(1+λx2)ddxγ+(γ2λγ)x2].\displaystyle\frac{1}{2}\left[-(1+\lambda x^{2})^{2}\frac{d^{2}}{dx^{2}}-2\lambda x(1+\lambda x^{2})\frac{d}{dx}-\gamma+(\gamma^{2}-\lambda\gamma)x^{2}\right].

By choosing γ=λ+λ2+42γ~\gamma=\frac{\lambda+\sqrt{\lambda^{2}+4}}{2}\equiv\tilde{\gamma}, for which (γ2λγ)=1(\gamma^{2}-\lambda\gamma)=1, we can get the following factorization form:

H~1=AA=Hγ2,\tilde{H}_{1}=A^{{\dagger}}A=H-\frac{\gamma}{2}, (19)

where HH is the Hamiltonian of the quantum oscillator (15). On the other hand, for the partner Hamiltonian H~2:=AA\tilde{H}_{2}:=AA^{{\dagger}} we find

H~2(γ)\displaystyle\tilde{H}_{2}(\gamma) =\displaystyle= A(γ)A(γ)\displaystyle A(\gamma)A^{{\dagger}}(\gamma)
=\displaystyle= 12[(1+λx2)2d2dx22λx(1+λx2)ddx+γ+(γ2+λγ)x2].\displaystyle\frac{1}{2}\left[-(1+\lambda x^{2})^{2}\frac{d^{2}}{dx^{2}}-2\lambda x(1+\lambda x^{2})\frac{d}{dx}+\gamma+(\gamma^{2}+\lambda\gamma)x^{2}\right].

Now, after straightforward calculations, the following relationship is hold:

H~2(γ)=H~1(γ1)+R(γ1),\displaystyle\tilde{H}_{2}(\gamma)=\tilde{H}_{1}(\gamma_{1})+R(\gamma_{1}), (21)

where γ1=γ+λ\gamma_{1}=\gamma+\lambda and R(γ)=γλ2R(\gamma)=\gamma-\frac{\lambda}{2}. As it is seen, the Hamiltonian H~1(γ)\tilde{H}_{1}(\gamma) admitting a factorization form A(γ)A(γ)A^{{\dagger}}(\gamma)A(\gamma) in such a way that the partner Hamiltonian H~2(γ)=A(γ)A(γ)\tilde{H}_{2}(\gamma)=A(\gamma)A^{{\dagger}}(\gamma) is of the form as H~1(γ)\tilde{H}_{1}(\gamma) but for a different value of the γ\gamma. In this case, it is usually said that there is shape invariance and makes it possible to calculate exact spectrum of the Hamiltonian [32, 33]. The ground state |ψ0(γ)|\psi_{0}(\gamma)\rangle with zero energy is given by A(γ)|ψ0(γ)=0A(\gamma)|\psi_{0}(\gamma)\rangle=0. Subsequently, by using Eq. (21) it is seen that |ψ0(γ1)|\psi_{0}(\gamma_{1})\rangle is an eigenstate of H~2(γ)\tilde{H}_{2}(\gamma) with the energy E~1=R(γ1)\tilde{E}_{1}=R(\gamma_{1}):

H~2(γ)|ψ0(γ1)=[H~1(γ1)+R(γ1)]|ψ0(γ1)=R(γ1)|ψ0(γ1).\tilde{H}_{2}(\gamma)|\psi_{0}(\gamma_{1})\rangle=[\tilde{H}_{1}(\gamma_{1})+R(\gamma_{1})]\ |\psi_{0}(\gamma_{1})\rangle=R(\gamma_{1})|\psi_{0}(\gamma_{1})\rangle. (22)

Similarity, using Eq. (2) we have

H~1(γ)[A(γ)|ψ0(γ1)]\displaystyle\tilde{H}_{1}(\gamma)[A^{{\dagger}}(\gamma)|\psi_{0}(\gamma_{1})\rangle] =\displaystyle= A(γ)H~2(γ)|ψ0(γ1)\displaystyle A^{{\dagger}}(\gamma)\tilde{H}_{2}(\gamma)|\psi_{0}(\gamma_{1})\rangle (23)
=\displaystyle= A(γ)[H~1(γ1)+R(γ1)]|ψ0(γ1)\displaystyle A^{{\dagger}}(\gamma)[\tilde{H}_{1}(\gamma_{1})+R(\gamma_{1})]\ |\psi_{0}(\gamma_{1})\rangle
=\displaystyle= R(γ1)[A(γ)|ψ0(γ1)].\displaystyle R(\gamma_{1})[A^{{\dagger}}(\gamma)|\psi_{0}(\gamma_{1})\rangle].

In other words, A(γ)|ψ0(γ1)A^{{\dagger}}(\gamma)|\psi_{0}(\gamma_{1})\rangle is the first excited state of H~1(γ)\tilde{H}_{1}(\gamma) with the energy E~1=R(γ1)\tilde{E}_{1}=R(\gamma_{1}). We can iterate this process and obtain the energy eigenvalues of H~1(γ)\tilde{H}_{1}(\gamma) as following:

E~n=k=1nR(γ+kλ)=n(γ+λ2n).\tilde{E}_{n}=\sum_{k=1}^{n}R(\gamma+k\lambda)=n(\gamma+\frac{\lambda}{2}n). (24)

Finally, by substituting γ=γ~=λ+λ2+42\gamma=\tilde{\gamma}=\frac{\lambda+\sqrt{\lambda^{2}+4}}{2} and using Eq. (19), we can obtain the eigenvalues EnE_{n} of the harmonic oscillator on the circle as

En=E~n+γ~2=γ~(n+12)+λ2n2.E_{n}=\tilde{E}_{n}+\frac{\tilde{\gamma}}{2}=\tilde{\gamma}(n+\frac{1}{2})+\frac{\lambda}{2}n^{2}. (25)

It is obvious that in the flat limit λ0\lambda\rightarrow 0, γ~1\tilde{\gamma}\rightarrow 1 and Eq. (25) reduces to the energy eigenvalues of the quantum harmonic oscillator on a flat line, i.e, En(λ=0)=n+12E_{n(\lambda=0)}=n+\frac{1}{2}.

3 Circular oscillator Hamiltonian as an ff-deformed quantum oscillator

The ff-deformed quantum oscillators [20], as the nonlinear oscillators are characterized by the following deformed dynamical variables A^\hat{A} and A^\hat{A}^{\dagger}

A^\displaystyle\hat{A} =\displaystyle= a^f(n^)=f(n^+1)a^,\displaystyle\hat{a}f(\hat{n})=f(\hat{n}+1)\hat{a},
A^\displaystyle\hat{A}^{\dagger} =\displaystyle= f(n^)a^=a^f(n^+1),\displaystyle f(\hat{n})\hat{a}^{\dagger}=\hat{a}^{\dagger}f(\hat{n}+1), (26)

where a^\hat{a} and a^\hat{a}^{\dagger} are usual bosonic annihilation and creation operators. The deformation function f(n^)f(\hat{n}) is an operator-valued function of the number operator n^(=a^a^)\hat{n}\ (=\hat{a}^{\dagger}\hat{a}) and the nonlinear properties of this system are governed by it. As usual, without loss of generality, we choose f(n)=f(n)f(n)=f^{{\dagger}}(n). From equation (3), it follows that the ff-deformed operators A^\hat{A}, A^\hat{A}^{\dagger} and n^\hat{n} satisfy the following closed algebra

[A^,A^]=(n^+1)f2(n^+1)n^f2(n^),\displaystyle[\hat{A},\hat{A}^{\dagger}]=(\hat{n}+1)f^{2}(\hat{n}+1)-\hat{n}f^{2}(\hat{n}),
[n^,A^]=A^,[n^,A^]=A^.\displaystyle[\hat{n},\hat{A}]=-\hat{A},\,\,\,\,\,\,\,\,\,[\hat{n},\hat{A}^{{\dagger}}]=\hat{A}^{{\dagger}}. (27)

The above-mentioned algebra represents a deformed Weyl-Heisenberg algebra whose nature depends on the nonlinear deformation function f(n^)f(\hat{n}). An ff-deformed oscillator is a nonlinear system characterized by a Hamiltonian of the harmonic oscillator form

H^=12(A^A^+A^A^).\hat{H}=\frac{1}{2}(\hat{A}^{\dagger}\hat{A}+\hat{A}\hat{A}^{\dagger}). (28)

Using Eq. (3) and the number state representation n^|n=n|n\hat{n}|n\rangle=n|n\rangle, the eigenvalues of the Hamiltonian (28) can be written as

En=12[(n+1)f2(n+1)+nf2(n)].E_{n}=\frac{1}{2}[(n+1)f^{2}(n+1)+nf^{2}(n)]. (29)

It is worth noting that in the limiting case f(n)1f(n)\rightarrow 1, the deformed algebra (3) and the deformed energy eigenvalues (29), respectively, will reduce to the standard Weyl-Heisenberg algebra and the harmonic oscillator spectrum.

Comparing the energy spectrum of the harmonic oscillator on a circular background, Eq. (25), with the energy spectrum of an ff-deformed oscillator, Eq. (29), we obtain the corresponding deformation function for the oscillator on the circle as

f(n^)=[γ~+(n^12)λ]1/2.f(\hat{n})=\left[\tilde{\gamma}+(\frac{\hat{n}-1}{2})\lambda\right]^{1/2}. (30)

Furthermore, the annihilation and creation operators of the harmonic oscillator on the circle can be written in terms of the conventional operators a^\hat{a} and a^\hat{a}^{\dagger} as follows

A^=a^[γ~+(n^12)λ]1/2,A^=[γ~+(n^12)λ]1/2a^.\hat{A}=\hat{a}\left[\tilde{\gamma}+(\frac{\hat{n}-1}{2})\lambda\right]^{1/2},\quad\quad\hat{A}^{\dagger}=\left[\tilde{\gamma}+(\frac{\hat{n}-1}{2})\lambda\right]^{1/2}\hat{a}^{\dagger}. (31)

These two operators satisfy the deformed Weyl-Heisenberg commutation relation

[A^,A^]=n^λ+γ~,[\hat{A},\hat{A}^{\dagger}]=\hat{n}\lambda+\tilde{\gamma}, (32)

and act upon the quantum number states |n|n\rangle, corresponding to the energy eigenvalues (25), as

A^|n\displaystyle\hat{A}|n\rangle =\displaystyle= f(n)n|n1,\displaystyle f(n)\sqrt{n}|n-1\rangle,
A^|n\displaystyle\hat{A}^{{\dagger}}|n\rangle =\displaystyle= f(n+1)n+1|n+1.\displaystyle f(n+1)\sqrt{n+1}|n+1\rangle. (33)

4 Nonlinear CSs on a circle

We can construct CSs for the ff-deformed oscillator similar to those of the harmonic oscillator. The nonlinear transformation of the creation and annihilation operators naturally provides the notion of nonlinear CSs or ff-coherent states. These states as eigenstates of the ff-deformed annihilation operator defined [20],

A^|z,f=z|z,f\hat{A}|z,f\rangle=z|z,f\rangle (34)

From Eq. (3), we can obtain an explicit form of the nonlinear coherent states on a circle (NCSsOC) in the number state representation as,

|z,f=𝒩(|z|2)1/2n=0znn!f(n)!|n,|z,f\rangle=\mathcal{N}(|z|^{2})^{-1/2}\sum_{n=0}^{\infty}\frac{z^{n}}{\sqrt{n!}f(n)!}|n\rangle, (35)

where by definition f(0)!=1f(0)!=1 and

f(n)!=f(n)f(n1)f(1),f(n)!=f(n)f(n-1)\cdots f(1), (36)

zz is a complex number, and the normalization constant 𝒩\mathcal{N} is given by

𝒩(|z|2)=n=0|z|2nn![f(n)!]2.\mathcal{N}(|z|^{2})=\sum_{n=0}^{\infty}\frac{|z|^{2n}}{n![f(n)!]^{2}}. (37)

Here, the deformation function corresponding to the oscillator on a circle f(n)!f(n)!, using Eq. (30) and after some calculations, is given by

f(n)!=(λ2)nΓ(β+n)Γ(β),f(n)!=(\frac{\lambda}{2})^{n}\frac{\Gamma(\beta+n)}{\Gamma(\beta)}, (38)

where β=2γ~λ+1\beta=\frac{2\tilde{\gamma}}{\lambda}+1 and Γ\Gamma is the gamma function. Similarly, we can define the f-deformed CSs corresponding to the harmonic oscillator on the circle as below:

|z,fcircle|zλ=𝒩(|z|2)1/2n=0znρ(n)|n,|z,f\rangle_{circle}\equiv|z\rangle_{\lambda}=\mathcal{N}(|z|^{2})^{-1/2}\sum_{n=0}^{\infty}\frac{{z}^{n}}{\sqrt{\rho(n)}}|n\rangle, (39)

where ρ(n)=n!(λ2)2n[Γ(β+n)Γ(β)]2\rho(n)=n!\ (\frac{\lambda}{2})^{2n}\left[\frac{\Gamma(\beta+n)}{\Gamma(\beta)}\right]^{2}. It is worth noting that to have states belonging to the Fock space, is required that 0<𝒩(|z|2)<0<\mathcal{N}(|z|^{2})<\infty, which implies that |z|limnn[f(n)]2=|z|\leq\lim_{n\rightarrow\infty}n[f(n)]^{2}=\infty , where in the last equality we have used Eq. (30). Moreover, it is obvious that in the flat limit, λ0\lambda\rightarrow 0, ρ(n)n!\rho(n)\rightarrow n! and the above deformed CSs reduce to the standard CSs:

|z=e|z|22n=0znn!|n.|z\rangle=e^{-\frac{|z|^{2}}{2}}\sum_{n=0}^{\infty}\frac{{z}^{n}}{\sqrt{n!}}|n\rangle. (40)

4.1 Resolution of identity

In this subsection, let us show the constructed NCSsOC form an overcomplete set. In other words, the following resolution of identity has to be satisfied

d2z|zλW(|z|2)λz|=n=0N|nn|=I.\displaystyle\int d^{2}z\ |z\rangle_{\lambda}\ W(|z|^{2})\ _{\lambda}\langle z|=\sum_{n=0}^{N}|n\rangle\langle n|=I. (41)

The resolution of identity can be achieved by finding a measure function W(|z|2)W(|z|^{2}). For this purpose, we substitute |zλ|z\rangle_{\lambda} from the Eq. (39) into Eq. (41), and yields

d2z|zλW(|z|2)λz|=πn=0|nn|ρ(n)d(|z|2)|z|2nW(|z|2)𝒩(|z|2),\displaystyle\int d^{2}z\ |z\rangle_{\lambda}\ W(|z|^{2})\ _{\lambda}\langle z|=\pi\sum_{n=0}^{\infty}\frac{|n\rangle\langle n|}{\rho(n)}\int d(|z|^{2})|z|^{2n}\frac{W(|z|^{2})}{\mathcal{N}(|z|^{2})}, (42)

where we have used: z=|z|eiθz=|z|{e^{i\theta}} and d2z=12d(|z|2)dθ{d^{2}}z=\frac{1}{2}d{(|z|^{2})}d\theta. By using the change of variable x=|z|2x=|z|^{2} and considering w(x)=πW(|z|2)𝒩(|z|2)w(x)=\pi\frac{W(|z|^{2})}{\mathcal{N}(|z|^{2})}, we have:

0w(x)xn𝑑x=ρ(n)=1[(2λ)2]nΓ(n+1)Γ2(β+n)Γ2(β).\int_{0}^{\infty}w(x){x^{n}}dx=\rho(n)=\frac{1}{[(\frac{2}{\lambda})^{2}]^{n}}\frac{\Gamma(n+1)\ \Gamma^{2}(\beta+n)}{\Gamma^{2}(\beta)}. (43)

The above integral is called the moment problem and well-known mathematical methods such as Mellin transformations can be used to solve it [34, 35]. From definition the of Meijers GG-function, it follows that,

𝑑xxk1Gp,qm,n(αx|a1,,an,an+1,,apb1,,bm,bm+1,,bq)\displaystyle\int dx\ x^{k-1}G_{p,q}^{m,n}\left(\alpha x\Big{|}\begin{array}[]{*{20}{c}}{a_{1},\cdots,a_{n},a_{n+1},\cdots,a_{p}}\\ {b_{1},\cdots,b_{m},b_{m+1},\cdots,b_{q}}\end{array}\right) (46)
=1αkj=1mΓ(bj+k)j=1nΓ(1ajk)j=m+1qΓ(1bjk)j=n+1pΓ(aj+k).\displaystyle=\frac{1}{{{\alpha^{k}}}}\frac{{\prod_{j=1}^{m}\Gamma({b_{j}}+k)\prod_{j=1}^{n}\Gamma(1-{a_{j}}-k)}}{{\prod_{j=m+1}^{q}}\Gamma(1-{b_{j}}-k){\prod_{j=n+1}^{p}}\Gamma({a_{j}}+k)}. (47)

Comparing Eq. (43) with Eq. (46), we find that the measure function can be written as,

w(x)\displaystyle w(x) =\displaystyle= 4λ2Γ2(β)G0,33,0(4xλ2|00,β1,β1).\displaystyle\frac{4}{\lambda^{2}\Gamma^{2}(\beta)}G_{0,3}^{3,0}\left(\frac{4x}{\lambda}^{2}\ \Big{|}\begin{array}[]{*{20}{c}}{0}\\ {0,\ \beta-1,\ \beta-1}\\ \end{array}\right). (50)

In this manner, it is seen that the NCSsOC satisfy the resolution of identity and consequently form an overcomplete set.

5 Quantum statistical properties of the NCSsOC

In this section, we turn to investigate some quantum optics features of the NCSsOC, such as probability of finding nn quanta, mean number of photons, the Mandel parameter and quadrature squeezing.

5.1 Photons-number distribution

Using Eq.(27), the mean number of photons in the NCSsOC is obtained as follows:

n^=λz|a^a^|zλ=n=0Nnp(n),\langle\hat{n}\rangle=\ _{\lambda}\langle z|{\hat{a}^{\dagger}}\hat{a}|z\rangle_{\lambda}=\sum_{n=0}^{N}np(n), (51)

where the probability of finding nn photon in the NCSsOC is given by

p(n)=|z|2n𝒩(|z|2)ρ(n).p(n)=\frac{|z|^{2n}}{\mathcal{N}(|z|^{2})\rho(n)}. (52)

As it is clear from the above equation, it is difficult to predict the results analytically. Thus, in Fig. 2 we show the effect of the parameters zz and λ\lambda on the probability of finding nn photons and also the mean number of photons in the NCSsOC.

Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 2: The probability of finding nn photon in the NCSsOC as a function of λ\lambda and zz (a,b). Here, z=3z=3 and λ=0.5\lambda=0.5 in panels (a) and (b), respectively. The mean number of photons in the NCSsOC versus λ\lambda and zz (c,d).

In Fig. 2(a) the probability p(n)p(n) of the NCSsOC for n>1n>1 is characterized by a fast initial increase followed by a very slow decrease and tends to zero, while the p(0)p(0) always enhances with increasing λ\lambda and eventually approaches unity, as expected from the conservation of probability. This behavior is also true for all other values of zz. Moreover, the probability of finding nn photon for the large value of nn is less than the NCSsOC with small photons. In Fig. 2(b), the probability p(n)p(n) for n>1n>1 show the same trend in Fig. 2(b). However, the probability p(0)p(0) decreases from its initial unit-value with increasing zz and eventually approaches zero. To get further insight, let us consider the limiting case λ0\lambda\rightarrow 0. From Eq. (52), we find that the probability p(n)p(n) tends to a Poissonian distribution, and the NCSsOC approaches the standard CSs.

The mean number of photons in the NCSsOC as a function of λ\lambda and zz are shown in Fig. 2(c) and  2(d), respectively. These results show that the highest value of the photons mean number occurs at the large (small) value of zz (λ\lambda). In other words, the mean number of photons in the NCSsOC is not significant when both parameters zz and λ\lambda are simultaneously small or large.

5.2 Mandel parameter

Refer to caption
Refer to caption
Refer to caption
Figure 3: The Mandel parameter of the NCSsOC as a function of λ\lambda and zz (a,b). (c) The density plot of Mandel parameter for the NCSsOC versus zz and λ\lambda.

In this subsection, we investigate deviation from the Poisson distribution for the NCSsOC using the Mandel parameter. This parameter is given by [36]:

Q=(Δn)2nn,Q=\frac{{{{(\Delta n)}^{2}}-\left\langle n\right\rangle}}{{\left\langle n\right\rangle}}, (53)

where the positive, zero, and negative values represent super-Poissonian, Poissonian, and sub-Poissonian distribution, respectively. Due to the complexity of the final form of the Mandel parameter for the NCSsOC, we do not attempt to obtain the analytic form. Instead, we numerically investigate the Mandel parameter for such states using Eqs. (51) and (52). In Fig. 3, we have plotted the Mandel parameter of the NCSsOC as a function of λ\lambda and zz. The results show that, for some fixed values of z=0.1,1,2z=0.1,1,2\, and 33, the Mandel parameter of the NCSsOC first becomes more sub-Poissonian with increasing λ\lambda and then tends to zero at large values of λ\lambda. Furthermore, this parameter is more negative with increasing zz (see Fig. 3 (a)). Whereas, for some fixed values of λ=0.01,1,2\lambda=0.01,1,2\, and 33, the Mandel parameter becomes more negative with increasing zz and then tends to 0.6-0.6 (1-1) at large (small) values of λ\lambda, as seen in Fig. 3 (b). We can justify this result by the fact that the NCSsOC behaves as the standard CSs at small λ\lambda and when the parameter zz is large enough. Indeed, the NCSsOC is the statndard CSs at exactly zero curvature, as shown in Eq. (40). In Fig. 3 (c), we observe that at the small value of λ\lambda the most negative amount of the Mandel parameter occurs in the NCSsOC with the small zz values.

5.3 Quadrature squeezing

Refer to caption
Refer to caption
Refer to caption
Refer to caption
Figure 4: The squeezing parameter S1S_{1} versus ϕ\phi for (a) z=1z=1, and (b) λ=0.5\lambda=0.5. Here, solid red, dashed blue, dashdotted orange and dotted green lines correspond to (a) ((b)) λ=0.01\lambda=0.01 (z=0.1)(z=0.1), λ=0.5\lambda=0.5 (z=0.5)(z=0.5), λ=1\lambda=1 (z=1)(z=1) and λ=1.5\lambda=1.5 (z=1.5)(z=1.5), respectively. The density plots of (c) S1S_{1}, and (d) S2S_{2} as functions of zz and λ\lambda for ϕ=π/6\phi=\pi/6.

In this subsection, we consider the quadrature operators X^1\hat{X}_{1} and X^2\hat{X}_{2} defined in terms of creation and annihilation operators a^\hat{a} and a^\hat{a}^{\dagger} as follows :

X^1=12(a^eiϕ+a^eiϕ),X^2=12i(a^eiϕa^eiϕ).\begin{array}[]{l}{{\hat{X}}_{1}}=\frac{1}{2}\left({\hat{a}{e^{i\phi}}+{{\hat{a}}^{\dagger}}{e^{-i\phi}}}\right),\\ {{\hat{X}}_{2}}=\frac{1}{2i}\left({\hat{a}{e^{i\phi}}-{{\hat{a}}^{\dagger}}{e^{-i\phi}}}\right).\end{array} (54)

By using the commutation relation of a^\hat{a} and a^\hat{a}^{\dagger}, the following uncertainty relation is obtained

(ΔX1)2(ΔX2)2116|[X^1,X^2]|2=116.{\left({\Delta{X_{1}}}\right)^{2}}{\left({\Delta{X_{2}}}\right)^{2}}\geq\frac{1}{{16}}{\left|{\left\langle{\left[{{{\hat{X}}_{1}},\left.{{{\hat{X}}_{2}}}\right]}\right.}\right\rangle}\right|^{2}}=\frac{1}{{16}}. (55)

As is known, the quadrature squeezing occurs if we have (ΔXi)2<1/4(i=1or2)(\Delta X_{i})^{2}<1/4(i=1or2) or equally Si4(ΔXi)21<0S_{i}\equiv 4{(\Delta X_{i})^{2}}-1<0. Employing Eqs. (51) and (54), after some algebra, the squeezing parameter S1S_{1} is obtained as

S1\displaystyle S_{1} =\displaystyle= 2cos2ϕz2𝒩(z2)n=2z2nn(n1)ρ(n2)ρ(n)(2cosϕz𝒩(z2)n=1z2nnρ(n1)ρ(n))2\displaystyle\frac{2\cos 2\phi}{z^{2}\mathcal{N}(z^{2})}\sum_{n=2}^{\infty}\frac{z^{2n}\sqrt{n(n-1)}}{\sqrt{\rho(n-2)\rho(n)}}-\Bigg{(}\frac{2\cos\phi}{z\mathcal{N}(z^{2})}\sum_{n=1}^{\infty}\frac{z^{2n}\sqrt{n}}{\sqrt{\rho(n-1)\rho(n)}}\Bigg{)}^{2} (56)
+2n^.\displaystyle+2\langle\hat{n}\rangle.

Here, zz is assumed to be a real parameter, and the mean number of photons, n^\langle\hat{n}\rangle, is already defined in Eq. (51). Figs. 4(a) and 4(b) display the squeezing parameter S1S_{1} for the NCSsOC as a function of φ\varphi for z=1z=1 and λ=0.5\lambda=0.5, respectively. The red solid, dashed blue, dashdotted orange and dotted green lines depict numerical results related to λ=0.01\lambda=0.01, λ=0.5\lambda=0.5, λ=1\lambda=1 and λ=1.5\lambda=1.5 for Fig. 4(a), and z=0.1z=0.1, z=0.5z=0.5, z=1z=1 and z=1.5z=1.5 for Fig. 4(b), respectively. These figures show a periodic dependant on ϕ\phi, and in areas that the NCSsOC is squeezed the amount of squeezing always enhances with increasing zz (see Fig. 4(b)). While, for the fixed value z=1z=1, the squeezing first enhances with increasing λ\lambda and then leads to decreasing squeezing, as seen in Fig. 4(a)). This behavior can be easily understood from Fig. 4(c) that at the small value of λ\lambda the highest amount of the squeezing appears in the NCSsOC with small zz, which is also in agreement with the results of the Mandel parameter. At λ=0\lambda=0, the squeezing properly vanished, as expected from the standard CSs. It is worth noting that the squeezing parameter S2S_{2} can be obtained from Eq. (56) by replacing ϕ=ϕ+π/2\phi=\phi+\pi/2. In this manner, the squeezing parameter S2S_{2} is plotted for different values of λ\lambda and zz in Fig. 4(d). This plot clearly exhibits the same behavior of the squeezing parameter S1S_{1} with ϕ=5π/6\phi=5\pi/6 (not shown here).

6 Summary and Concluding Remarks

In this paper, we have obtained the Hamiltonian of a harmonic oscillator confined on a circle using the gnomonic projection. It is shown that the algebra of such harmonic oscillator can be regarded as ff-deformed harmonic oscillator algebra. We have constructed the nonlinear CSs for such harmonic oscillator so that the NCSsOC properly tends to the standard CSs at zero curvature. We then studied the quantum statistical properties of the NCSsOC, and found that the nonclassical features of these states enhance with increasing the parameter zz even in small curvatures of the circle.

7 Acknowledgment

A.M. wishes to thank The Office of Graduate Studies and Research Vice President of The University of Isfahan for their support. E.A. also wishes to thank the Shahrekord University for their support.

References

  • [1] R. Shankar, Principles of Quantum Mechanics, Springer, 1994.
  • [2] P. F. Bracken, Int. J. Theor. Phys. 46, 116 (2007).
  • [3] V. V. Gritsev and Y. A. Kurochkin, Phys. Rev. B 64, 035308 (2001).
  • [4] D. V. Bulaev, V. A. Geyler, and V. A. Margulis, Phys. Rev. B 69, 195313 (2004).
  • [5] J. P. Gazeau and W. Piechocki, J. Phys. A: Math. Gen. 37, 6977 (2004).
  • [6] J. F. Carinena, M. F. Ranada, and M. Santander, Ann. Phys. 322, 2249 (2007).
  • [7] J. F. Carinena, M. F. Ranada, and M. Santander, J. Math. Phys. 52, 072104 (2011).
  • [8] P. Bracken, J. Math. Phys. 55, 102102 (2014).
  • [9] H. Jensen and H. Koppe, Ann. Phys. 63, 586 (1971).
  • [10] R. C. T. da Costa, Phy. Rev. A 23, 1982 (1981).
  • [11] R. C. T. da Costa, Phys. Rev. A 25, 2893 (1982).
  • [12] Sh. Dehdashti, R. Roknizadeh, A. Mahdifar, and H. Chen H, Int. J. Theor. Phys. 55, 124 (2016).
  • [13] P. W. Higgs, J. Phys. A: Math. Gen. 12, 309 (1979).
  • [14] H. I. Leemon, J. Phys. A: Math. Gen. 12, 489 (1979).
  • [15] R. J. Glauber, Phys. Rev. 136, 2529 (1963); R. J. Glauber, Phys. Rev. 131, 2766 (1963); R. J. Glauber, Phys. Rev. Lett. 10, 84 (1963).
  • [16] A.P. Perelomov, Generalized Coherent States and their Applications, Springer, 1989.
  • [17] S. T. Ali, J.P. Antoine, J. P. Gazeau, Coherent States, Wavelets and their Generalizations, Springer, 2000.
  • [18] J. R. Klauder, B. S. Skagerstam, Coherent States, Applications in Physics and Mathematical Physics, World Scientific, 1985.
  • [19] A. I. Solomon, Phys. Lett. A 196, 29 (1994); J. Katriel and A. I. Solomon, Phys. Rev. A 49, 5149 (1994); P. Shanta, S. Chaturvedi, V. Srinivasan, and R. Jagannathan, J. Phys. A 27, 6433 (1994).
  • [20] V. I. Man’ko, G. Marmo, E. C. G. Sudarshan, and F. Zaccaria, Phys. Scr. 55, 528 (1997).
  • [21] W. Vogel, R.L. de Matos Filho, Phys. Rev. A 52 4214 (1995).
  • [22] R. L. de Matos Filho, W. Vogel, Phys. Rev. A 54 4560 (1996).
  • [23] A. Mahdifar, W. Vogel, Th. Richter, R. Roknizadeh, and M. H. Naderi, Phys. Rev. A 78, 063814 (2008).
  • [24] M. H. Naderi, M. Soltanolkotabi, and R. Roknizadeh, Eur. Phys. J. D 32, 397 (2005).
  • [25] A. Mahdifar, R. Roknizadeh and M. H. Naderi, Int. J. of Geom. Meth. in Mod. Phys. 9, 1250009 (2012).
  • [26] V. N. Popov, New J. Phys. 6, 17 (2004).
  • [27] A. Mahdifar, R. Roknizadeh, M.H. Naderi, J. Phys. A 39, 70037014 (2006).
  • [28] A. Mahdifar, T. Hoseinzadeh, M. Bagheri Harouni, Annals of Physics 355, 21 (2015).
  • [29] H. Bagheri and A. Mahdifar 54, 052104 (2013).
  • [30] E. Amooghorban and A. Mahdifar, Annals of Physics 360, 237 (2015).
  • [31] A. Mahdifar and E. Amooghorban, J. Math. Phys. 60, 082106 (2019).
  • [32] B. K. Bagchi, Supersymmetry In Quantum and Classical Mechanics, Chapman and Hall/CRC, 2000.
  • [33] J. F. Cariñena, M. F. Rañada, M. Santander, Rep. Math. Phys., 54, 285 (2004).
  • [34] J. R. Klauder, K. A. Peson, J. M. Sixdeniers, Phys. Rev. A 64, 013817 (2001).
  • [35] A. M. Mathai, R. K. Saxena, Generalized Hypergeometric Functions with Applica-tions in Statistics and Physical Sciences, Springer-Verlag, 1973.
  • [36] L. Mandel and E. Wolf, Optical Coherence and Quantum Optics, Cambridge University Press, Cambridge, 1995.