1.
Introduction
The quantum calculus, briefly called q-calculus, is infact the investigation of calculus without limits. The q-calculus is a generalization of ordinary calculus as for q→1−, the quantum calculus reduces to the ordinary calculus. Recently, it has got attention of many researchers in the field of applied mathematics due to the fact that it provides a connection between mathematics and physics. It has been proved very helpful in the mathematical modelling of the problems arising in quantum computing. This motivates the study of certain special functions in the context of q-calculus. The q-analogues of some special functions like q-Gamma and q-Beta functions, 1-variable q-Hermite polynomials and q-Hermite polynomials-Appell polynomials etc., are introduced and studied [5,6,15,17,19].
In this paper, we introduce the 2-variable q-Hermite polynomials and investigate some of its properties like series definition, recurrence relations, q-differential equation, summation formulas, operational definition and some integral representations.
We review briefly some definitions and notations of the quantum calculus.
The q-analogue of a complex number α is defined as [2,3,11,12]:
The first five few q-numbers are as follows:
The q-factorial is defined as:
which satisfies
The Gauss q-binomial coefficient is defined as [2,3,11,12]:
The raising and lowering q-powers are defined as [2,3,11,12]:
or, equivalently
where [nk]q is given by Eq (1.3). For n=1, it is obvious that (u±a)1q=(u±a).
In particular, for a=0, Eq (1.4) or (1.5), gives
The two q-exponential functions, denoted by eq(u) and Eq(u), are defined as [2,3,11,12]:
and
respectively. The relation between both q-exponential functions is as follows:
Next, the q-derivative of a function f with respect to u, denoted by Dq,uf(u), is defined as [13]:
Also, for two arbitrary functions g(u) and h(u), we have
In particular, we have
and
The mth order q-derivative of the q-exponential functions are as follows:
and
where Dmq,u denotes the mth order q-derivative with respect to u.
Also, the q-definite integral of a function g is defined as [11,12,14]:
and
From Eqs (1.15) and (1.16), it is clear that
The q-definite integral of the q-derivative of a function g(u) is given as:
We note that, for q→1−, all the results in the q-calculus reduce to the corresponding results in ordinary calculus.
The Hermite polynomials are one of the most applicable classical orthogonal special functions. They are the solutions of the differential equations, which are equivalent to the Schrödinger equation for a harmonic oscillator in quantum mechanics and thus they appear as eigen functions. Moreover, these polynomials have significant role in the study of classical boundary-value problems in parabolic regions, through the use of parabolic coordinates, or in quantum mechanics as well as in other application areas.
Now, we recall the generating function and series definition of the classical Hermite polynomials Hn(u) and the 2-variable Hermite polynomials Hn(u,v).
The classical Hermite polynomials Hn(u) are defined by means of the following generating function [1]:
and the series definition [1]:
The 2-variable Hermite polynomials (2VHP) Hn(u,v) are defined by means of the following generating function [4]:
and the series definition [4]:
The pure and the differential recurrence relations for Hn(u,v) are defined as [4]:
and
The 2VHP Hn(u,v) satisfies the following differential equation [4]:
From Eqs (1.19) and (1.21), it is clear that the 2VHP Hn(u,v) is related with the classical Hermite polynomials Hn(u) as:
The q-Hermite polynomials of 1-variable, which is defined in several ways, has many literatures due to its wide applications in various fields of mathematics and physics. For instance, Berg and Ismail [6], have shown that 1-variable q-Hermite polynomials can be used to build the classical q-orthogonal polynomials systemetically.
Recently, Nalci and Pashaev [17] defined the q-Hermite polynomials by means of the following generating function :
and series definition
in the context of q-analog of shock soliton solution.
In this paper, we introduce the 2-variable q-Hermite polynomials and study certain properties of these polynomials. An operational definition and some integral representations of these polynomials are obtained. Some examples are also considerd to show the efficacy of the proposed method. Some concluding remarks are given. At the end of this paper the graphical representations of these polynomials of certain degrees with specified values of q are given.
In the next section, we introduce 2-variable q-Hermite polynomials with the help of generating function and we obtain their series definition, recurrence relations, operational differential equations of 2VqHP.
2.
The 2-variable q-Hermite polynomials
In this section, we introduce the 2-variable q-Hermite polynomials and obtain their series definition, recurrence relations and differential equation.
In view of Eqs (1.21), (1.29) and (1.30), we define the 2-variable q-Hermite polynomials (2VqHP) Hn,q(u,v) by means of the following generating function:
The series on the right hand side of the above equation converges in the finite complex plane.
Expanding the left hand side of Eq (2.1) by using Eq (1.7), we have
which on using the following series rearrangement technique[1]:
gives
Equating the coefficients of equal powers of t from both sides of the above equation, we have the following series definition of 2VqHP Hn,q(u,v):
where [.] denotes the greatest integer function. Being a finite power series, the series on the right hand side of the above equation converges in the finite complex plane.
Taking v=0 and u=0 one by one in Eq (2.3), we have the following boundary conditions:
Further, from Eq (2.3), it can be easily verified that
where a is a constant.
Now, taking q-partial derivative of both the sides of Eq (2.1) with respect to u and then using Eq (1.13) in the resultant equation, we have
Using Eq (2.1) in the left hand side of Eq (2.6), we have
or, equivalently
Comparing the equal powers of t from both the sides of the above equation, we have
Again, taking 2nd order q-partial derivative of both the sides of Eq (2.1) with respect to u, then using Eq (1.13) for m=2 and following the steps involved in the proof of Eq (2.7), we have
Similarly, taking mth order q-partial derivative of both sides of Eq (2.1) with respect to u, then using Eq (1.13) and following the same steps, we get the following mth order q-partial derivative of Hn,q(u,v) with respect to u:
Next, taking q-partial derivative of both sides of Eq (2.1) with respect to v and following the steps involved in the proof of Eq (2.7), we have
Similarly, taking mth order q-partial derivative of both sides of Eq (2.1) with respect to v and following the same steps, we have
Again, in view of Eqs (2.7), (2.10) and (2.5), it is easy to verify that the 2VqHP Hn,q(u,v) satisfy the following q-partial differential equation:
Next, we proceed to establish the recurrence relations for the 2VqHP Hn,q(u,v). For this, we need to prove the following lemma:
Lemma 2.1. If Dq,t denotes the q-partial derivative with respect to t, then
Proof. Using Eqs (1.7) and (1.13), we have
The series on the right hand side of the above equation converges in the finite complex plane.
Since, from Eq (1.1), we have [2n]q[n]q = 1+qn, therefore using Eq (1.2) in the right hand side of the above equation, we have
which on simplifying the right hand side, gives
Using Eq (1.7) in the right hand side of above equation, we get the assertion (2.13).
Remark 2.1 Since, for q→1−, we have Dq,t → Dt and eq(vt2) → evt2, therefore for q→1−, Lemma 2.2, gives the following well known result of ordinary calculus:
where Dt denotes the ordinary derivative with respect to t.
Now, we establish the pure and q-differential recurrence relations for Hn,q(u,v) in the form of following theorem:
Theorem 2.2. The 2VqHP Hn,q(u,v) satisfy the following recurrence relations:
and
Proof. Taking q-derivative of both sides of Eq (2.1) with respect to t and then using Eq (1.11) in the right hand side of the resultant equation, we have
which on using Eqs (1.12) and (1.2) in the left hand side and using Eqs (1.13) and (2.13) in the right hand side, gives
Using Eq (2.1) in both sides of the above equation and then comparing the equal powers of t from both sides of the resultant equation, we get the assertion (2.16).
If a is an arbitrary constant, then in view of Eq (2.5), we have
which on using Eq (2.7), gives
Again, using Eq (2.5) in the right hand side of Eq (2.21), we have
Using Eq (2.22) for a=1 and a=q in the right hand side of Eq (2.16), we have the assertion (2.17).
Now, replacing n by n−1 in Eq (2.17), we have
which on using Eq (2.7) in the right hand side, gives the assertion (2.18).
Next, replacing n by n−1 in Eq (2.16), we have
In view of Eq (2.5), for any constant a, we have
which on using Eq (2.10), gives
Again, using Eq (2.5) in the right hand side of Eq (2.24), we have
Using Eq (2.25) for a=1 and a=q, in the right hand side of Eq (2.23), we have the assertion (2.19).
Again, replacing n by n−1 in Eq (2.19), we have
which on using Eq (2.10) in the right hand side of Eq (2.26), gives the assertion (2.20).
In order, to establish the differential equation of 2VqHP Hn,q(u,v), we define the following operators:
Let f(u,v) be a q-function of two variables, then we define the shift operators La,u and La,v as:
and
where a is a constant. In view of (2.27), the shift operator La,u satisfy the following properties:
In particular, for a = b, we have
If L−1a,u is the inverse of the operator La,u, that is L−1a,uLa,u=I, where I is an identity operator such that If(u,v)=f(u,v).
Then, from Eq (2.27), we have
Replacing au by u in the above equation, we have
which on using Eq (2.27), gives
Using induction method, Eqs (2.29) and (2.30), gives
where r is any integer.
Similarly, it can be shown that La,v satisfies the following properties:
where r is any integer.
Now, we prove the following result:
Theorem 2.3. The 2-variable q-Hermite polynomials Hn,q(u,v) satisfy the following q-differential equation:
where L.,u is the shift operator defined in Eq (2.27).
Proof. Replacing n by (n−1) in Eq (2.16) and then using (2.7) in resultant equation, we have
In view of Eq (2.27), we have
Also, in view of Eqs (2.5) and (2.27), we have
Using Eqs (2.27), (2.37) and (2.38) in Eq (2.36), we have
Now, using Eqs (2.7) and (2.8) in Eq (2.39), we have
or, equivalently
which, on further simplification yields assertion (2.35).
Remark 2.2 It can be easily verified that
Using Eq (2.41), for m=1 and m=2, in Eq (2.40), we have the following equivalent form of differential equation:
Another, forms of the differential Eq (2.35) are as follows
and
where Lq,v is the shift operator defined by Eq (2.28).
Remark 2.3 Since, for q→1−, the q-calculus reduces to the ordinary calculus. Therefore, for q→1−, Eq (2.1) gives the generating function of 2VHP Hn(u,v), given by Eq (1.21). Also, the series definition, recurrence relations and differential equation for 2VHP Hn(u,v), given by the Eqs (1.21), (1.22) and (1.23)-(1.28), can be obtained by taking q→1− in Eqs (2.3), (2.16)-(2.20) and (2.35), respectively.
In the next section, we establish certain summation formulas involving the 2-variables q-Hermite polynomials Hn,q(u,v) and its q-derivatives.
3.
Summation formulas
In this section, we obtain certain summation formulas for 2VqHP by exploiting the identities (1.7) and (1.9) and using the generating function of Hn,q(u,v). The following summation formulas for the 2VqHP Hn,q(u,v) hold:
Theorem 3.1. The following summation formulas hold in the finite complex plane:
If n is even i.e. n=2m (m∈N), then
if n is odd i.e. n=2m+1 (m∈N∪{0}), then
Proof. In view of Eq (1.9), we have
which on using Eqs (1.7), (1.8) and (2.1), gives
or, equivalently
Using Eq (2.2) in the left hand side of above equation, we have
which on equating the coefficients of equal powers of t from both sides, gives the assertion (3.1).
Again, using Eq (1.9), we have
Using Eqs (1.7), (1.8) and (2.1) in the above equation, we have
which on using the following series rearrangement technique[1]:
gives
Comparing the even and odd powers of t from both sides of Eq (3.7), we get the assertions (3.2) and (3.3), respectively.
Remark 3.1 The summation formula, given by Eq (3.3), has the following equivalent form:
From Theorem 3.1 and Remark 3.1, we deduce the following summation formulas for q-derivative of 2VqHP Hn,q(u,v) with respect to u:
Corollary 3.2. The following summation formulas hold in the finite complex plane:
and
Proof. Using Eq (2.7), we have
which on simplifying, gives
Using Eqs (3.13) and (1.6) in Eq (3.1), we have the assertion (3.9).
Similarly, using Eq (2.7), we have
Using Eqs (3.14) and (1.6) in Eq (3.2), we have the assertion (3.10).
From Eq (3.14), we have
Using Eq (3.15) in Eq (3.8), we have the assertion (3.11).
Remark 3.2 Using Eq (2.10) in the right hand sides of Eqs (3.1), (3.2) and (3.8), we have the following summation formulas for q-derivative of 2VqHP Hn,q(u,v) with respect to v:
Corollary 3.3. The following summation formulas hold in the finite complex plane:
and
Remark 3.3 For q→1−, Eqs (3.1)-(3.3), (3.8)-(3.11), (3.16)-(3.18) give summation formulas for 2VHP Hn(u,v) and its derivatives.
Now, we list some examples in Table 1 to show the efficacy of our results established in sections 2 and 3.
In the next section, we obtain the operational and integral representations for the 2VqHP Hn,q(u,v).
4.
Operational and integral representations
It has been realized that the use of operational identities have simplified the study of special polynomials. In this section, we obtain the operational and integral representations of 2VqHP Hn,q(u,v).
First, we establish the following result:
Theorem 4.1. The operational representation of 2-variable q-Hermite polynomials Hn,q(u,v) is as follows:
Proof. In view of Eq (1.12), we have
Using the above equation in the right hand side of Eq (2.3), we have
which on using Eq (1.6), gives
Using Eq (1.7) in the right hand side of above equation, we have the assertion (4.1).
Remark 4.1 For q→1−, Eq (4.1) becomes [4]:
Now, we obtain the following integral representations for the 2VqHP Hn,q(u,v):
Theorem 4.2. The definite q-integral of 2VqHP Hn,q(u,v) with respect to u is as follows:
Proof. Using Eq (2.7), we have
which on using Eq (1.18) in the right hand side, gives the assertion (4.4).
Next, we establish the following result:
Theorem 4.3. The definite q-integral of 2VqHP Hn,q(u,v) with respect to v is as follows:
Proof. Using Eq (2.10), we have
which on using Eq (1.18) in the right hand side, gives assertion (4.5).
In view of Theorems 4.2 and 4.3, we have the following result:
Corollary 4.4. The double q-integration of 2VqHP Hn,q(u,v) is as follows:
Proof. Integrating Eq (4.4) with respect to v by taking limit from c to d and using Eq (1.18), we have
which in view of Eq (4.5), gives
Similarly, integrating Eq (4.5) with respect to u by taking limit from a to b and using Eq (1.18), we have
which in view of Eq (4.4), gives
In view of Eqs (4.7) and (4.9), we have the assertion (4.6).
Remark 4.2 For q→1−, Eqs (4.4) and (4.5) reduce to the integral representations for 2VHP Hn(u,v) [4]. Also, for q→1−, Eq (4.6) gives:
Examples:
Now, we consider some examples to show the efficacy of the results obtained, in this section.
I. First, we take n=6 in Eqs (4.1), (4.4)-(4.6) to obtain the corresponding results for H6,q(u,v).
For n=6, Eq (4.1), the following operational representation of H6,q(u,v):
Also, taking n=6,a=1,b=2,c=2,d=3 in Eqs (4.4)-(4.6) and then simplifying by using Eq (2.3), we have the following integral representations of H6,q(u,v):
and
II. Next, we take n=3 in Eqs (4.1), (4.4)-(4.6) to obtain the corresponding results for H3,q(u,v).
For n=3, Eq (4.1), the following operational representation of H3,q(u,v):
Also, taking n=3,a=0,b=2,c=3,d=4 in Eqs (4.4)-(4.6) and then simplifying by using Eq (2.3), we have the following integral representations of H3,q(u,v):
and
5.
Concluding remarks
The post quantum calculus, briefly called (p,q)-calculus, is considered as a generalization of the q-calculus by some mathematicians, as for p=1, the (p,q)-calculus reduces to the q-calculus. The (p,q)-analogues of certain ordinary special functions and q-special functions as well as polynomials like Beta function, Gamma function, Euler polynomials and Bernoulli polynomials have been introduced and studied [5,9,16,18].
We conclude this paper with the introduction and study of (p,q)-analogue of 2VqHP Hn,q(u,v).
We recall the (p,q)-number [n]p,q, which is defined as [8]:
for any positive integer n∈N. The (p,q)-factorial is defined as [8]:
Also, we recall that
The (p,q)-exponential function, denoted by ep,q(u), is defined as [8]:
The (p,q)-derivative of a function f with respect to u, denoted by Dp,q,u f(u), is defined as [7]:
Recently, Duran et al. defined the 1-variable (p,q)-Hermite polynomials as [10]:
In view of Eqs (2.1) and (5.5), we define the 2-variable (p,q)-Hermite polynomials 2V(p,q)HP Hn,p,q(u,v) by means of the following generating function:
The series on the right hand side of the above equation converges in the finite complex plane.\
Simplifying and equating the coefficients of equal powers of t from both sides of Eq (5.6), gives the following series definition of 2V(p,q)HP Hn,p,q(u,v):
Being a finite power series, the series on the right hand side of the above equation converges in the finite complex plane.
Similar to the Lemma 2.1, we find the following (p,q)-partial derivative of the function ep,q(vt2):
which is used to obtain the following (p,q)-pure and differential recurrence relations for 2V(p,q)HP Hn,p,q(u,v):
and
Now, in order to establish the operational representation for the 2V(p,q)HP Hn,p,q(u,v), we define a (p,q)-operator Xp,q,u as:
which, in view of Eqs (2.31), (2.32) and (5.2), acts on a (p,q)-function in the following manner:
where L−1p,u is the inverse shift operator, defined by Eq (2.31). It is easy to verify that the shift-differential operator Xp,q,u satisfies the following:
Using the above equation in the right hand side of Eq (5.7), we have
which on using Eq (5.2), gives
Using Eq (5.3) in the right hand side of above equation, we have the following operational identity of the 2-variable (p,q)-Hermite polynomials Hn,p,q(u,v):
where the (p,q)-operator X2p,q,u:=p−1L−2p,uD2p,q,u.
Acknowledgments
The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work through Research Group Program under Grant No. RGP. 2/51/42.
Conflict of interest
The authors declare that they have no competing interest.
Supplementary
In order to plot the graphs of the 2-variable q-Hermite and (p,q)-Hermite polynomials for certain values of p and q, we obtain the following first few q-Hermite and (p,q)-Hermite polynomials by using Eqs (2.3) and (5.7) respectively:
H0,q(u,v)=1,
H1,q(x,y)=u,
H2,q(u,v)=u2+[2]qv,
H3,q(u,v)=u3+[3]q[2]quv,
H4,q(u,v)=u4+[4]q[3]qu2v+[4]q[3]qv2
and
H0,p,q(u,v)=1,
H1,p,q(u,v)=u,
H2,p,q(u,v)=pu2+[2]p,qv,
H3,p,q(u,v)=p3u3+[3]p,q[2]p,quv,
H4,p,q(u,v)=p6u4+[4]p,q[3]p,qp2u2v+[4]p,q[3]p,qpv2.
The following graphical representations of the first few 2-variable q-Hermite polynomials Hn,q(u,v) and 2-variable (p,q)-Hermite polynomials Hn,p,q(u,v) for certain values of p and q, are obtained by using the above suitable expression in the software MATLAB: