Arxiv:1312.6571V1 [Math-Ph] 23 Dec 2013 Spr Mgnr Ignl Oevr N Etfnto Spa Function Test Any Which Moreover, for Product, Diagonal

arXiv:1312.6571v1 [math-ph] 23 Dec 2013 spr mgnr ignl oevr n etfnto spa function test any which Moreover, for product, diagonal. Wick imaginary the pure under is not but product Weyl-Moyal ed r enda prtrvle itiuin.Tespa The distributions. operator-valued as defined are fields eraigfntos swl nw,teuulaimtcqu axiomatic usual the space known, Schwartz prod well these the As of each i under functions. role algebras important decreasing are An that spaces zero. function quantization. is by part the symmetric specifies the antisymme part product whose Moyal symmetric matrix whose constant a and by tensor defined of products form star of specific the prop of those independent explore are and ther that define algebras to literature useful product the is formulations it Wick-Voros in a [9]–[11], and Since theory product Weyl-Moyal conve the star absolute of a condition. of equivalence of property this the use satisfy having the and models through impo cone formulated light is be the account with can into QFT shown products NC As for star [5]–[8]. models the noncommutative of of interpretation (NC nature theory nonlocal field quantum the noncommutative pr in Wick t significance and new Weyl-Moyal in a the considered for previously specifically was operator and quantization products corresponding their star the consider of precisely algebras, convergence describe these and of spaces, properties convergen basic absolutely are some products and the defining series power orsodne ikodrn,Fc-aganrepresenta Fock-Bargmann ordering, Wick correspondence, LERSWT OVRETSA RDCSADTHEIR AND PRODUCTS STAR CONVERGENT WITH ALGEBRAS ecnie ieetqatztosi nfigwy starti way, unifying a in quantizations different consider We fun analytic of algebras product star study we paper, this In 2000 e od n phrases. and words Key ahmtc ujc Classification. Subject Mathematics n ai rpriso hs lers hc r sflfor algebras. useful th operator of are corresponding representation their which space of Hilbert algebras, description the to these given of is presen attention and properties way basic unifying a and alge in qua products Such noncommutative star in different consider absolutely. and theory converge quantization products deformation the defining series Abstract. ERSNAIN NHLETSPACES HILBERT IN REPRESENTATIONS esuysa rdc lerso nltcfntosfrwhi for functions analytic of algebras product star study We S falrpdydcesn mohfntos nwihquantu which on functions, smooth decreasing rapidly all of eomto uniain oa trpout wse con twisted product, star Moyal quantization, deformation 1. .A SOLOVIEV A. M. Introduction 35,8S0 66,4H5 63,46N50. 46H30, 46H15, 46L65, 81S30, 53D55, 1 tion. .Mt.Phys. Math. J. lersadt h exact the to and algebras e eut ntestructure the on results t ce riso h ovretstar convergent the of erties .W ics h structure the discuss We t. rsaientrlyin naturally arise bras n[] aslt condition causality a [6], in lers h usinof question The algebras. tmfil hoy We theory. field ntum ecneto deformation of context he csadcnito rapidly of consist and ucts Hilbert in representations sn osnu nthe on consensus no is e plctos Special applications. dcs[][] tacquires It [1]–[4]. oducts S h trproduct. star the ntecs fteWeyl- the of case the In simple some and rgence, fnnomttv field noncommutative of u prahi played is approach our n nu edter uses theory field antum ehvn h structure the having ce gfo eea form general a from ng sa ler ne the under algebra an is rcpr stePoisson the is part tric tn o h physical the for rtant F) eas taking because QFT), tosfrwihthe which for ctions h ymti part symmetric the FIAN-TD-2012-02 htepower the ch 54 gbaassociated lgebra 757(2013) 073517 , ouin Weyl volution, m 2 M. A. SOLOVIEV of an algebra under the Wick product must consist of entire analytic functions. This suggests that any attempt to develop a Wick-Voros formulation of noncommutative QFT should be based on treating it as an essentially nonlocal theory, which can also be confirmed by the simple but instructive example of the normal ordered star product square of a free field, considered for the Weyl-Moyal and Wick products in [8] and [12]. We believe that using appropriate function algebras consisting of rapidly decreasing functions and universal for different star products is very useful in the general theory of quantization on symplectic linear spaces. In particular, their operator representations associated with a representation of the translation group are defined in an obvious way and can be extended by continuity to larger algebras. In [3], a faithful representation of a Wick star product algebra with absolute convergence is defined by means of the Gelfand-Naimark-Segal construction. We construct Hilbert space representations of various star product algebras explicitly and characterize the domain of definition of the corresponding operator algebras. The paper is organized as follows. In Sec. 2, we introduce the main definitions and notation. In Sec. 3, we consider the spaces ρ consisting of entire functions of order ρ 2 and of minimal type, and prove that theseE spaces are topological algebras under ≤ any star product ⋆ϑ defined by a constant matrix ϑ. In Sec. 4, we show that the algebra ρ ( , ⋆ϑ) contains a family of subalgebras associated naturally with cone-shaped regions. E ρ ρ If ρ > 1, then ( , ⋆ϑ) has a subalgebra W consisting of functions rapidly decreasing at real infinity.E These are just the functions that can serve as test functions for NC QFT, and we explain the relation of W ρ with the Gelfand-Shilov spaces Sβ used in the previous papers [6], [8], and [12]. In Sec. 5, we describe the Fourier transform of the ρ algebra (W , ⋆ϑ), which is used later in Sec. 6 to construct Hilbert space representations ρ of this algebra. In Sec. 7, we prove that the representations of (W , ⋆ϑ) can be uniquely ρ extended by continuity to representations of the algebra ( , ⋆ϑ). We also characterize exactly the domain of definition of the corresponding operatE or algebra in the space L2 of square integrable functions. Analogous results for the Fock-Bargmann space representation are derived in Sec. 8. The final Sec. 9 contains concluding remarks.

2. Star products on symplectic linear spaces Let X be be a real 2n-dimensional vector space equipped with a symplectic bilinear form ω. Choosing a symplectic basis in X, we identify the form ω with its matrix 0 I Ω= n and we denote by π the inverse matrix of Ω. Then the Poisson bracket In 0 of two− functions on X is written as

2n ∂f ∂g f, g = πjk . { } ∂xj ∂xk j,kX=1

The Weyl-Moyal star product f ⋆~ g is a noncommutative deformation of the ordinary (pointwise) product of the functions in the direction of the Poisson bracket and is deﬁned by

i~ ←−∂ πjk −→∂ (f ⋆~ g)(x)= f(x) e 2 j k g(x), (1) where the Planck constant ~ plays the role of a noncommutativity parameter, and where the summation convention for the repeated indices is used. In this paper, we consider ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 3 star products of the more general form

jk i~ ←−∂j ϑ −→∂k (f ⋆~ϑ g)(x)= f(x) e g(x)=

~ m m ∞ (i ) jk = f(x)g(x)+ ϑ ∂ j ∂ ′k f(x)g(x′) , (2) m! x x ′ mX=1 x=x where ϑ is a constant and, in general, complex matrix. By the classical-quantum corres- 1 pondence principle, the antisymmetric part of ϑ must be equal to π. Then, and only 2 then, the product (2) satisﬁes the limit relation 1 lim (f ⋆~ ϑ g g ⋆~ ϑ f)= f, g . ~ 0 i~ − { } → Therefore, in what follows we assume that

1 1 jk kj ϑ = Q + π, π =Ω− , Q = Q . (3) 2 Different choices of the symmetric matrix Q correspond to different quantizations, i.e., to different rules of association between functions on X and quantum operators, includ- ing the standard, the anti-standard, the normal, and the anti-normal ordering rules. This is demonstrated below by constructing explicitly Hilbert space representations which convert the star products to the operator product. In Sec. 3–5 we write ⋆ϑ instead of ⋆~ϑ, assuming that the deformation parameter ~ is included in the matrix ϑ. We note that the general form (2) of star product was used in recent papers by Blaszak and Domański [13] and Omori et al. [14], but the attention of these papers was focused on other aspects. The first of them presents the formalism of the phase space quantum mechanics in a very systematic way, but does not consider problems with convergence of the infinite series representing the star products, whereas the second paper focuses mainly on the important case of the ⋆-exponential functions of quadratic forms and their related Clifford algebra and does not concern the representation of star product algebras in Hilbert spaces.

3. Convergence of the star products We will consider spaces of functions for which the star product (2) is absolutely convergent for any matrix ϑ. Such algebras were previously studied [1], [2] mostly for the case of Weyl-Moyal product. The generalization to arbitrary star products in not diﬃcult, but we give it for completeness in the form convenient for deriving our main results in Secs. 4–8. Let 0 <ρ< and let ρ(Rd) denote the space of all smooth functions on Rd satisfying the inequalities∞ E κ κ 1 1/ρ x/L ρ ∂ f(x) C L−| |(κ!) − e| | , (4) | |≤ L where L> 0 and can be chosen arbitrarily large, CL is a constant depending on f and Zd L, x = max xj , and κ = (κ1,...,κd) +. Here and hereafter, we use the standard | | 1 j d | | ∈ multi-index≤ ≤ notation κ κ ∂| | ∂ = κ1 κd , κ = κ1 + + κd, κ!= κ1! ...κd!. ∂x1 ...∂xd | | · · · The topology of ρ(Rd) is deﬁned by the family of norms E κ κ L ∂ f(x) ρ f = sup | | e x/L , L> 0. (5) L | 1 1/ρ | −| | k k x,κ (κ!) − 4 M. A. SOLOVIEV

κ κ1 κd We note that κ! in (4), (5) can be replaced with κ = κ1 ...κd , because by Stirling’s formula we have κ κ κ κ κ e−| |κ κ! C (1 + δ)| |e−| |κ for any δ > 0. (6) ≤ ≤ δ However the definition in form (4), (5) is more convenient for what follows. The definition of ρ(Rd) can be reformulated in terms of complex variables. Let E ρ(Cd) be the space ofE all entire functions of order ρ and minimal type, i.e. satisfying the condition ≤ ǫ z ρ f(z) C e | | , (7) | |≤ ǫ where ǫ> 0 and can be chosen arbitrarily small, z = x + iy, and z = max zj . The | | 1 j d | | ≤ ≤ natural topology on E ρ(Cd) is defined by the countable system of norms ǫ z ρ 1 1 f ǫ = sup f(z) e− | | , ǫ = 1, 2 , 3 ,.... (8) k k z | | and is hence metrizable. It follows from the elementary properties of analytic functions of several variables that E ρ(Cd) is complete and therefore a Fréchet space. Proposition 1. The space E ρ(Cd) is canonically isomorphic to ρ(Rd). E Proof. The Taylor series expansion of f ρ(Rd) about a point x Rd has an infinite radius of convergence and defines an analytic∈E continuation of f to C∈d which satisfies (7). Indeed, using the first of inequalities (6) and choosing L′

κ κ κ /ρ κ! ǫ 2z ρ ∂ f(z) f ǫ 2| |(ǫρe)| | e | | . (12) | | ≤ k k (κκ)1/ρ Using the second of inequalities (6), this can also be written as κ κ κ /ρ 1 1/ρ ǫ 2z ρ ∂ f(z) C ′ f 2| |(ǫ′ρ)| | (k!) − e | | for any ǫ′ > ǫ. (13) | |≤ ǫ k kǫ ρ d We infer that the restriction f Rd of the function to the real space belongs to (R ) and satisﬁes | E 1 1 f Rd L C′ f ǫ for L< min 1, , k | k ≤ ǫ,Lk k 2ǫ1/ρ ρ1/ρ which completes the proof. ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 5

Remark 1. We prefer to say that E ρ(Cd) is isomorphic to ρ(Rd) instead of saying that these spaces coincide, because the elements of ρ(Rd) canE also be continued as anti-analytic functions. E We now characterize the infinite order differential operators that are endomorphisms of ρ(Rd). E κ Theorem 1. Let ρ> 1 and ρ′ = ρ/(ρ 1). If G(s)= κ cκs is an entire function of − κ ρ Rd order ρ′ and finite type, then the differential operatorPG(∂)= κ cκ∂ maps ( ) ≤ κ E continuously into itself and the series κ cκ∂ f converges absolutelyP in every norm of this space for any f ρ(Rd). If 0

ǫ and can be taken arbitrarily close to ǫ. Because ǫ can be chosen arbitrarily small, we conclude that the series c ∂κf converges absolutely in E ρ(Cd) ρ(Rd) κ κ ≈E and the operator G(∂) is continuousP in the topology of this space. If ρ 1, then it suffices to take into account that the Taylor coefficients of any entire function≤ satisfy κ the inequality cκ Cεε| | with arbitrarily small ε > 0. Combining this inequality with (13), we see| that| ≤ in this case, any operator of the form G(∂), with G(s) an entire function, maps ρ(Rd) continuously into itself. The theorem is proved. E Proposition 2. The set of polynomials is dense in E ρ(Cd). Proof. We show that the Taylor series expansion of f E ρ(Cd) converges absolutely in E ρ(Cd). Let ǫ > 2ρdǫ. Using the inequality d z ρ ∈ z ρ together with (11) and ′ | | ≥ j | j| (12), we get P κ κ ′ ρ z κ ∂ f(0) κ (ǫ /d) P zj ∂ f(0) | | sup z e− j | | κ! ′ ≤ κ! | | ≤ Xκ ǫ Xκ z κ κ /ρ f 2| |(ǫd/ǫ′)| | C ′ f . ≤ k kǫ ≤ ǫ,ǫ k kǫ Xκ Because the topology of E ρ(Cd) is finer than the topology of simple convergence, we conclude that the Taylor series converges absolutely in the topology of E ρ(Cd) to the function f. The proposition is proved. Theorem 2. If ρ 2, then for any matrix ϑ, the space ρ(Rd) is an associative ≤ E unital topological algebra under the star product f ⋆ϑ g. The series defining this product converges absolutely in ρ(Rd). E Proof. According to the definition (2), the product f ⋆ϑ g is obtained by applying the jk ρ 2d operator exp(iϑ ∂j ∂k) to the function (f g)(x,x′) (R ) and then restricting ⊗ ⊗ jk ∈E to the diagonal x = x′. The entire function exp(iϑ sjsk′ ) is of order 2 and finite type. From the isomorphism ρ(Rd) E ρ(Cd), we immediately infer that the restriction map E ≈ 6 M. A. SOLOVIEV

ρ(R2d) ρ(Rd) is continuous. Therefore Theorem 1 implies that the series deﬁning E →E ρ d ρ d f ⋆ϑ g is absolutely convergent. The bilinear map (f, g) f g from (R ) (R ) ρ 2d → ⊗ E ×E to (R ) is obviously jointly continuous. Therefore the bilinear map (f, g) f ⋆ϑ g is alsoE jointly continuous. To prove the associativity of this product we use the→ identity κ κ ′ ∂x f(x,x) = (∂x + ∂x ) f(x,x′) x=x′ , (15) which holds for any f ρ(R2d). With the help of (15) we obtain ∈E jk jk iϑ ∂ j (∂ ′k +∂ ′′k ) iϑ ∂ ′j ∂ ′′k (f ⋆ϑ (g ⋆ϑ h))(x)= e x x x f(x) e x x g(x′)h(x′′) (16) x=x′=x′′

jk jk iϑ (∂ j +∂ ′j )∂ ′′k iϑ ∂ j ∂ ′k ((f ⋆ϑ g) ⋆ϑ h)(x)= e x x x e x x f(x)g(x′) h(x′′) (17) x=x′=x′′

The right-hand sides of (16) and (17) coincide by Theorem 1 applied to the operator jk jk ρ R3d exp(iϑ ∂xj (∂x′k + ∂x′′k )+ iϑ ∂x′j ∂x′′k ) acting to ( ). Clearly, the function f(x) 1 is the unit element of the algebra ( ρ(Rd), ⋆ ).E The theorem is proved. ≡ E ϑ 4. Subalgebras associated with cone-shaped regions If ρ > 1, then the space ρ(Rd) has a nontrivial subspace W ρ(Rd) consisting of functions rapidly decreasing atE real infinity. More precisely, W ρ(Rd) consists of smooth functions on Rd satisfying the condition κ κ 1 1/ρ N ∂ f(x) C L−| |(κ!) − (1 + x )− for all L> 0,N = 0, 1, 2,.... (18) | |≤ N,L | | Correspondingly, its topology of is defined by the system of norms κ (1 1/ρ) N κ f N,L = sup L| |(κ!)− − (1 + x ) ∂ f(x) , (19) k k x,κ | | | | and under this topology W ρ(Rd) is a Fréchet space. By reasoning similar to the proof of Preposition 1, this space is canonically isomorphic to the space of entire functions satisfying the inequalities N ǫ y ρ (1 + x ) f(x + iy) C e | | , ǫ> 0,N = 0, 1, 2,..., (20) | | | |≤ N,ǫ and the system of norms (19) is equivalent to the system N ǫ y ρ f N,ǫ = sup(1 + x ) f(x + iy) e− | | . (21) k k x,y | | | | A similar space can be associated with each open cone V Rd. We define W ρ(V ) to be the space of all entire functions with the finite norms ⊂ N ǫδρ (x) ǫ y ρ f V,N,ǫ = sup(1 + x ) f(x + iy) e− V − | | , (22) k k x,y | | | | where δV (x) = infξ V x ξ is the distance of x from V . We note that δV (tx)= δV (x) ∈ | − | for all t > 0, because cones are invariant under dilations. Clearly, if V1 V2, then ρ ρ ρ d ⊂ W (V2) is canonically embedded in W (V1). The space W (R ) is embedded in every W ρ(V ) and all of these spaces are embedded in ρ(Rd). If K is a closed cone and a ρ d E ρ functional v (W (R ))′ has a continuous extension to each W (V ), where V K 0 , then the cone∈ K can be thought of as a carrier of v, because this property implies⊃ \{ that} v decreases in all directions outside K more rapidly than any test function. Theorem 3. If 1 ρ 2, then for any open cone V Rd, the space W ρ(V ) is an ≤ ≤ ⊂ associative topological algebra under the star product f ⋆ϑ g and the series defining this product converges absolutely in W ρ(V ). ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 7

Proof. Let f, g W ρ(V ). It is easy to see that f g W ρ(V V ) and ∈ ⊗ ∈ × f g V V,N,2ǫ f V,N,ǫ g V,N,ǫ, k ⊗ k × ≤ k k k k ρ if y in (22) is deﬁned by y = maxj yj . It is clear that if h(z, z′) W (V V ), then | | | | | | ρ ∈ × the function hˇ(z) = h(z, z) belongs to W (V ) and hˇ V,N,ǫ h V V,N,ǫ. Therefore, it suﬃces to prove an analogue of Theorem 1 for Wk ρk(V ).≤ To k thisk × end we derive an estimate similar to (13). Using Cauchy’s inequality and the elementary inequality (1 + x ) (1 + ξ )(1 + x ξ ), we obtain | | ≤ | | | − |

N κ κ κ /ρ 1 1/ρ ǫδρ (2x)+ǫ 2y ρ (1 + x ) ∂ f(z) C ′ f 2| |(2ǫ′ρ)| | (k!) − e V | | , (23) | | | |≤ N,ǫ k kV,N,ǫ κ which holds for any ǫ′ >ǫ. We now can state that if G(s)= κ cκs is an entire function κ of order ρ′ = ρ/(ρ 1) and ﬁnite type, then the diﬀerentialP operator G(∂)= κ cκ∂ W ≤ρ − κ maps (V ) continuously into itself and the series κ cκ∂ f converges absolutelyP in ρ this space. Indeed, if ǫ′′ > 2 ǫ, then it follows from (14)P and (23) that

κ N κ ǫ′′δρ (x) ǫ′′ y ρ c ∂ f ′′ c sup (1 + x ) ∂ f(z) e− V − | | k κ kV,N,ǫ ≡ | κ| | | | | Xκ Xκ z=x+iy κ κ /ρ κ /ρ′ C ′ f 2| |(2ǫ′ρ)| | (aρ′)| | , ≤ ǫ k kV,N,ǫ Xκ where ǫ′ > ǫ and can be taken arbitrarily close to ǫ. In particular, the operator jk ρ exp(iϑ ∂j ∂k) is a continuous automorphism of W (V V ), and so the theorem is proved. ⊗ ×

Remark 2. The family of subalgebras W 2(V ) of the maximal algebra 2(Rd) with the absolute convergence property plays a special role. In [6] and [12], we usedE the notation S 1/2 instead of W 2, because this space can be regarded as the projective limit of the Gelfand-Shilov spaces S1/2,b as b 0. The notation W 2 is here more convenient and → β,b 1/(1 β),b agrees with the notation in [15], where the isomorphism between S and W − is W ρ ρ,b established for β < 1. The spaces = b 0 W should be distinguished from the ρ ρ,b → S 1/2 W 2 spaces W = b W considered by GelfandT and Shilov [15]. The space = was proposedS in→∞ [6] as a universal test function space for noncommutative quantum ﬁeld theory. It was argued there that the spaces S 1/2(V ) associated with cones can be used to formulate a generalized causality condition and shown that some simple noncommutative models satisfy this condition.

5. Deformed convolution product ρ We now turn to describing the Fourier transform of the algebra (W , ⋆ϑ). As in the case [15] of the spaces W ρ, the behavior of the Fourier transforms of the elements of W ρ is characterized by an indicator function which is the Young dual of yρ. 8 M. A. SOLOVIEV

Let, as before, ρ > 1 and let ρ′ be the dual exponent of ρ, i.e., 1/ρ′ + 1/ρ = 1. We d d define Wρ′ (R ) to be the space of all smooth functions on R with the finite norms ′ κ a σ ρ g N,a = max sup ∂ g(σ)) e | | , a> 0,N = 0, 1, 2,.... (24) k k κ N σ | | | |≤ j jk In what follows we use the notation σ x = j σjx , σ ϑt = j σjϑ tj. The Fourier d/·2 iσ x · transform is defined by fˆ(σ) = (2π)− e−P· f(x)dx. P R d Theorem 4. If ρ′ 2, then for any complex matrix ϑ, the space Wρ′ (R ) is an associative topological algebra≥ under the operation g , g g ⊛ g , where 1 2 → 1 ϑ 2 1 i(σ τ) ϑτ (g1 ⊛ϑ g2)(σ)= g1(σ τ)g2(τ)e− − · dτ = (2π)d/2 Z −

1 iτ ϑ(σ τ) g1(τ)g2(σ τ)e− · − dτ. (25) (2π)d/2 Z − ρ d d The Fourier transformation is an isomorphism of (W (R ), ⋆ϑ) onto (Wρ′ (R ), ⊛ϑ), where ρ = ρ(ρ 1). ′ − Proof. We note that the bilinear operation deﬁned by (25) is a noncommutative defor- d/2 mation of the ordinary convolution (up to a factor (2π)− ). In the case when ϑ is a real skewsymmetric matrix, it is called the twisted convolution, and this case corresponds to the Weyl-Moyal star product. For the properties of the twisted convolution product and its relation to the Weyl-Heisenberg group, we refer the reader to [16] and [17]. Let us now show that the Laplace transformation

d/2 iσ (x+iy) g f(x + iy) = (2π)− e · g(σ)dσ (26) → Z d ρ d maps W ′ (R ) onto W (R ). Under the condition ρ 2, the integral in (26) is well ρ ′ ≥ deﬁned and analytic in Cd, and we have the chain of inequalities

N κ iσ (x+iy) κ (1 + x ) f(x + iy) CN max x f(x + iy) CN′ max e · ∂ g(σ) dσ | | | |≤ κ N | |≤ κ N Z | | | ≤ | ≤ ′ ρ′ ′ ρ′ σ y (d+1)a σ σ y a Pj σj CN′ g N,(d+1)a′ e− · − | | dσ CN,a′ g N,(d+1)a′ sup e− · − | | . ≤ k k Z ≤ k k σ A standard calculation gives

ρ′ ρ ρ sup  σ y a′ σj  = a yj ad y , σ − · − | | | | ≤ | | Xj Xj   where ρ ρ′ (a′ρ′) (aρ) = 1. (27) W ρ Rd We infer that f ( ) and f N,ǫ CN,a′ g N,(d+1)a′ for ǫ ad. Conversely, let f ρ ∈d k k ≤ d/2k k iσ (x+iy) ≥ belong to W (R ). Then the integral (2π)− e− · f(x + iy) dx is independent of y. Denoting it by g(σ) and assuming that κR N, we have | |≤ κ d/2 σ y κ ∂ g(σ) (2π)− inf e · (x + iy) f(x + iy) dx | |≤ y Z | | d/2 σ y N N (2π)− inf e · (1 + y ) (1 + x ) f(x + iy) dx ≤ y Z | | | | | | ρ ′ ρ′ σ y+a P yj a σ CN,a f N+d+1,ǫ inf e · j | | CN,a f N+d+1,ǫe− | | , ≤ k k y ≤ k k ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 9 where a>ǫ can be taken arbitrarily close to ǫ, and a′ and ρ′ are deﬁned by (27). We d ρ d d see that g Wρ′ (R ) and the map f g = fˆ from W (R ) to Wρ′ (R ) is continuous. ∈ → jk The Fourier transformation converts the operator exp(iϑ ∂j ∂k) into the multi- jk ⊗ plication by the function exp iσjϑ τk which is obviously a pointwise multiplier of 2d {− } 2d W ′ (R ) for ρ 2. Let h(σ, τ) be an arbitrary function in W ′ (R ). The formula ρ ′ ≥ ρ 1 1 1 ′ h(σ τ,τ)dτ eiσ xdσ = h(σ, τ)eiσ x+iτ x dσdτ d/2 d/2 · d · · (2π) Z (2π) Z − (2π) Z ′ x=x shows that the operation of restriction to the diagonal is converted by the Fourier transformation into the operation

2d d 1 Wρ′ (R ) Wρ′ (R ): h(σ, τ) h(σ τ,τ)dτ. → → (2π)d/2 Z − Consequently, \ f1 ⋆ϑ f2 = fˆ1 ⊛ϑ fˆ2, (28) which completes the proof.

6. Operator representations of star products Let T be a unitary projective representation of the translation group of a symplectic space (X,ω) and let exp i ω be its multiplier, so that we have the relation { 2~ } i ′ ~ ω(x,x ) TxTx′ = e 2 Tx+x′ . (29) We let H denote the Hilbert space on which this representation acts. The Weyl quantization associates Hilbert space operators with functions on X R2n in the following way ≈ 1 f f = f˜(s) T~sds, (30) 7−→ (2π)n Z where f˜ is the symplectic Fourier transform of f, deﬁned by 1 f˜(s)= f(x)eiω(x,s)dx. (31) (2π)n Z Letting, as before, Ω denote the matrix of the symplectic form ω, we have ω(x,s) = j k x Ωs = x Ωjks . Then f˜(s) = fˆ( Ωs), and after the change of variables σ = Ωs, the· formula (28) can be written as − −

^ t f1 ⋆ϑ f2 = f˜1 ⊛ f˜2, where ϑ =Ω ϑ Ω= Ω ϑ Ω. (32) ϑe − The integral in (30) is well deﬁned if f˜ is integrable,e and the Weyl transform takes the star product (1) into the operator product. Let now the matrix ϑ be of the form given by (3), and consequently, 1 ϑ = Q Ω, where Q = ΩQΩ. (33) − 2 − Then the Weyl correspondencee e (30) should be replacede by i~ 1 s Qs f f = f˜(s) e 2 · e T~sds. (34) 7−→ (2π)n Z Theorem 5. If 1 ρ 2, then for any ϑ of the form (3), the map (34) is a contin- ≤ ≤ ρ 2n uous homomorphism of the algebra (W (R ), ⋆~ϑ) into the algebra (H) of bounded operators on the Hilbert space H. B 10 M. A. SOLOVIEV

Proof. It follows from Theorem 4 and from the invariance of W ρ(R2n) under linear change of variables that the symplectic Fourier transformation is a topological and ρ 2n 2n W R ~ W ′ R ⊛ algebraic isomorphism of ( ( ), ⋆ ϑ) onto ( ρ ( ), ~ϑ). Because ρ′ 2, we have e ≥ ~ 1 Q s 2 f f˜(s) e 2 | e|·| | ds Ca f˜ 0,a, k k≤ (2π)n Z | | ≤ k k where a> (~/2) Q . It remains to show that the map | | i~ e 2n ˜ 1 ˜ s Qs Wρ′ (R ) (H): f f = f(s) e 2 · e T~sds −→ B 7−→ (2π)n Z is an algebra homomorphism. Using (29), (33), the symmetry of the matrix Q, and the Fubini theorem, we obtain e i~ 1 ˜ s Qs f = (f1 ⊛~ f2)(s) e 2 · e T~sds = (2π)n Z ϑe i~ 1 i~ t ϑ(s t)+ s Qs = f˜1(t)f˜2(s t)e− ·e − 2 · e T~sdtds = (2π)2n ZZ − i~ 1 i~ t ϑt′+ (t+t′)Q(t+t′) = f˜1(t)f˜2(t′)e− ·e 2 e T~ ′ dtdt′ = (2π)2n ZZ (t+t ) i~ i~ 1 t Qt+ t′ Qt′ = f˜1(t)f˜2(t′)e 2 · e 2 · e T~tT~t′ dtdt′ = f1f2. (2π)2n ZZ The theorem is proved.

In quantum mechanics on phase space, the following notation is usually used: x = 1 n j (p1,...,pn; q ,...,q ), where q are the coordinate variables and pj are their conjugate momentums. The standard projective representation of the phase space translations in the Hilbert space L2(Rn) is written as

i ~ (pq qp) Tp,q = e − , (35)

j i where q is the operator of multiplication by the jth coordinate function and ~ pj is the operator of differentiation with respect to the same coordinate. Let s = (u, v) and 1 0 I Q = n . Then s Qs = u v and the correspondence (34) takes the form 2 In 0 · · e e ~ 1 i u v+i(uq vp) 1 iuq ivp f f = f˜(u, v)e 2 · − dudv = f˜(u, v)e e− dudv, (36) 7−→ (2π)n ZZ (2π)n ZZ which defines the so-called qp-quantization, i.e., the standard ordering of the operators. We let jst denote the map defined by (36). In the theory of pseudodifferential operators, it is usually called the Kohn-Nirenberg correspondence. Under this correspondence the monomial pκqλ transforms into the operator qλpκ. Formally, this follows from the relations

κ λ n κ λ κ λ κ λ iuq ivp κ λ λ iuq κ ivp p q = (2π) ( i)| |i |∂ δ(v)∂ δ(u), ∂ ∂ e e− = ( i)| |i |q e p e , − v u − but ag rigorous proof is based on the extension of map (36) to tempered distributions [18] An alternative proof, based on extending it to the space 2, is given in Sec. 7. The star product corresponding to the standard ordering is writtenE as

i~ ∂←−p ∂−→j (f ⋆st g)(p,q)= f(p,q) e− j q g(p,q). (37) ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 11

1 0 I Let us point out some other important cases. The matrix Q = n deﬁnes −2 In 0 the correspondence e

1 ivp iuq f f = f˜(u, v)e− e dudv, 7−→ (2π)n ZZ under which the monomial pκqλ transforms to pκqλ, and so we have the anti-standard 2 i ς− In 0 ordering. Now, let Q = 2 , where the dimensional parameter ς is such −2 0 ς In 1 that ςq and ς− p aree of the same dimension. In this case, it is convenient to use the holomorphic variables

j 1 j 1 1 j 1 z = (ςq + iς− pj), z¯j = (ςq iς− pj). (38) √2 √2 − By 1, the elements of ρ(R2n) and W ρ(R2n) can be regarded as entire functions of these variables, consideredE as independent variables. The restriction to the real space R2n amounts then to identifyingz ¯ with the complex conjugate of z. With these variables, the formula (31) becomes

1 z w¯ z¯ w f˜(w, w¯)= f(z, z¯)e · − · dz dz,¯ (2πi)n ZZ ∧ 1 where w = (ςu iς− v)/√2. As usual, we introduce the annihilation and creation operators − j 1 j 1 1 j 1 a = (ςq + iς− pj), a† = (ςq iς− pj). √2 j √2 − Then the representation operator (35) can be written as

1 † 1 1 † 1 ~ (za z¯a) ~ zz¯ ~ za ~ z¯a Tz = e − = e− 2 e e− . In terms of the new variables, the quadratic form s Qs becomes iw w¯, and the correspondence (34) takes the form · − · e 1 wa† w¯a f f = f˜(w, w¯) e e− dw dw.¯ (39) 7−→ (2πi)n ZZ ∧ Therefore, in this case we have the Wick quantization, or in other words, the normal κ λ κ λ ordering, under which the monomialz ¯ z corresponds to the operator (a†) a . The corresponding star product is

~ ←−∂z −→∂z¯ (f ⋆Wick g)(z, z¯)= f(z, z¯)e · g(z, z¯). By changing the sign of Q, we obtain the quadratic form iw w¯, which defines the anti-Wick quantization and the anti-normal ordering aλ(a )κ. · e † 7. Extensions of representations We now show that the map (34) can be naturally extended to functions in ρ(R2n), where ρ 2. For clarity we first consider the representation in the space ofE square integrable≤ functions L2(Rn). In doing so we use another function space whose elements decrease at infinity faster than the elements of ρ increase. γ n E Let γ > 1 and ρ > 1. The space Wρ (R ) consists of all entire analytic functions satisfying the condition a q ρ+b y γ ψ(q + iy) Ce− | | | | (40) | |≤ where a, b, and C are positive constants depending on ψ. As before, we really deal with the restrictions of these functions to Rn, using the analytic continuation as an auxiliary 12 M. A. SOLOVIEV

γ 1 1/γ − tool. The space Wρ coincides with the Gelfand-Shilov space S1/ρ and is nontrivial if and only if γ ρ. Its natural topology is the inductive limit topology with respect ≥ γ,b to the family of Banach spaces Wρ,a equipped with the norms

a q ρ b y γ ψ a,b = sup ψ(q + iy) e | | − | | . (41) k k q,y | | γ Using reasoning similar to that in the proof of Theorem 4, it is easy to show that Wρ is invariant under the Fourier transformation if and only if 1/γ + 1/ρ = 1. It is important γ 2 2 γ that Wρ is dense in L . Indeed, if f L , ϕ Wρ and ϕ(q)dq = 1, then the sequence n ∈ γ∈ 2 f (q) = ν f(q ξ)ϕ(νξ)dξ belongs to Wρ and convergesR to f in the norm of L as ν − ν . R →∞ γ 2 We let Wρ (L ) denote the algebra of operators which are defined on the subspace γ A 2 γ Wρ of the Hilbert space L and map Wρ continuously into itself, and we endow this γ algebra with the topology of uniform convergence on bounded subsets of Wρ . Theorem 6. If 1 < ρ 2, then the map (34) defined in Theorem 5 as an algebra homomorphism from W ρ(≤R2n) to (L2) extends uniquely to a continuous homomorphism ρ R2n B 2 of the algebra ( ( ), ⋆~ϑ) to the algebra ρ′ (L ), where ρ′ = ρ/(ρ 1). E AWρ − Proof. Let f(p,q) ρ(R2n). This function can be written as ∈E κ λ f(p,q)= cκ,λp q , (42) Xκ,λ 1 where by the estimate (12) the coefficients c = ∂κ∂λf(0, 0) satisfy κ,λ κ!λ! p q ρ ( κ + λ )/ρ (2 ǫρe) | | | | c f ǫ . (43) | κ,λ| ≤ k k (κκλλ)1/ρ We first extend the transformation (36) that defines the standard ordering. If such a continuous extension to ρ exists, then it is unique because W ρ is dense in ρ by 2. Indeed, every polynomialEP (z) can be approximated in the topology of E ρ by a sequenceE ρ of functions of the form fν(z) = P (z)f0(z/ν), where f0 W is such that f0(0) = 1. ρ′ n ∈ Let ψ Wρ (R ). We define an operator f by ∈ def κ λ κ (fψ)(q) = c ( i~)| |q ∂ ψ(q) (44) κ,λ − Xκ,λ

ρ′ and show that the series on the right-hand side of (44) converges to an element of Wρ , ρ′ and f is hence well deﬁned as an operator from Wρ ito itself. To this end, we estimate the derivatives of ψ in the complex domain, using the Cauchy inequality again. Let z = q + iy. Taking into account that ρ ρ, we get ′ ≥ ′ ′ κ κ! κ! a q/2 ρ+b 2y ρ a ξ ρ+b 2η ρ ∂ ψ(z) κ sup ψ(z ζ) κ ψ a,be− | | | | sup e | | | | | |≤ r ζ=ξ+iη Dr | − |≤ r k k ξ+iη Dr ≤ ∈ ∈ ′ ′ ρ′ κ! a q/2 ρ+b 2y ρ a+(a+2ρ b) P r ψ e− | | | | e j j . ≤ rκ k ka,b

Using (11) with ρ′ instead of ρ, we obtain

It follows from (43) and (45) that

ρ′ n if ǫ is sufficiently small compared to a and b. Therefore, fψ Wρ (R ). The operator ρ′ ∈ f maps Wρ into itself continuously, because the right-hand side of (46) contains the ρ factor ψ a,b. Moreover, the Kohn-Nirenberg map jst : f f is continuous from to k2 k → E ρ′ (L ), because the right-hand side of (46) contains f ǫ. AWρ k k We now show that if f W ρ(R2n), then the definition of f by (44) is equivalent to the above definition (36). To∈ accomplish this, we consider the matrix element ϕ, f ψ , ρ′ n 2 hn i where ϕ, ψ Wρ (R ) and the angle brackets denote the inner product of L (R ). The function ∈ iuq ivp iuq χ(u, v)= ϕ, e e− ψ = ϕ¯(q)e ψ(q ~v)dq h i Z − ρ′ 2n ρ′ 2n belongs to Wρ (R ), becauseϕ ¯ ψ Wρ (R ) and this space is invariant under linear changes of variables and under the⊗ partial∈ Fourier transform. From (36), we have 1 1 ϕ, f ψ = f˜(u, v)χ(u, v)dudv = f(p,q)˜χ( p, q)dpdq = h i (2π)n ZZ (2π)n ZZ − − 1 κ λ = cκλ p q χ˜( p, q)dpdq, (2π)n ZZ − − Xκ,λ where the order of summation and integration can be interchanged because of absolute convergence. Taking the (inverse) symplectic Fourier transform, we obtain

κ λ κ λ κ λ κ ϕ, f ψ = cκλ( i)| |i | ∂ ∂ δ, χ = cκλ( i~)| | ϕ¯(q)q ∂ ψ(q)dq. h i − v u − Z Xκ,λ Xκ,λ Thus, the definitions (36) and (44) are indeed consistent. The star product (37) is continuous in ρ by Theorem 2, and the operator product is separately continuous in 2 E ρ ρ′ (L ) by the definition of the topology of bounded convergence. Because W is AWρ dense in ρ, we conclude that the constructed extension of correspondence (36) is an E ρ 2 algebra homomorphism from ( , ⋆~ϑ) to ρ′ (L ). E AWρ 1 0 In We let Qst denote the matrix defining the map jst and let jQ denote −2 In 0 the symbol-operator correspondence of general form (34). If f W ρ, then for any Q, jk ∈ s Qs˜ Q ∂j ∂ f˜(s)e · is the symplectic Fourier transform of e− k f. Hence we have the relation

i~ jk jk (Qst Q )∂j ∂ j (f)= j e 2 − k f , (47) Q st 14 M. A. SOLOVIEV

ρ ρ 2 which holds for all f W . We can now extend jQ to a map from into ρ′ (L ) ∈ E AWρ by replacing jst in (47) with its extension constructed above. The map deﬁned in this way is continuous, because the diﬀerential operator on the right-hand side of (47) is a continuous automorphism of ρ by Theorem 1. This completes the proof. E 8. The Fock-Bargmann space representation The Wick correspondence is usually realized using a representation in the Fock- Bargmann space F 2, which consists of antiholomorphic functions on Cn satisfying the condition 1 2 z 2 ϕ(¯z) e− ~ k k dµ(z) < , Z | | ∞ where dµ(z) = (2πi) ndz z¯ = π nd(Re z)d(Im z) and z 2 = z z¯ = z 2. It is a − ∧ − k k · j | j| Hilbert space with inner product given by P 1 2 1 z ϕ, ψ = ϕ(¯z)ψ(¯z)e− ~ k k dµ(z). h i ~n Z The representation (35) on the space L2 can be transferred to a representation on F 2 by means of the Bargmann transformation

1 2 1 2 def n/4 ~ z¯ ~ (√2z ¯ q) ψ(q) (Bψ)(¯z) = (π~)− e 2 e− 2 − ψ(q)dq, (48) −→ Z which is a unitary isomorphism from L2 onto F 2 (see [18], [19]). When realized on F 2, the creation operator is the multiplication byz ¯ and the annihilation operator is ~ ∂z¯. Accordingly, the representation of the phase-space translations is realized by the operators 1 1 1 1 ~ wz¯ w∂¯ z¯ ~ ww¯ ~ wz¯ w∂¯ z¯ BT B− = = e − = e− 2 e e− , w Tw and the Wick correspondence takes the form

wz¯ ~w∂¯ z¯ f f = f˜(w, w¯) e e− dµ(w), (49) 7−→ Z 2,A Let 1 < ρ 2 and A> 0. We deﬁne E¯ρ to be the space of all entire antiholomorphic functions satisfying≤ the condition A z 2 r z ρ ϕ(¯z) Ce k k − k k , (50) | |≤ 2,A where C and r are constants depending on ϕ, and we give E¯ρ the topology of the inductive limit of Banach spaces with the norms A z 2+r z ρ ϕ r = sup ϕ(¯z) e− k k k k . (51) k k z | |

Theorem 7. Suppose 1 < ρ 2 and ρ′ = ρ/(ρ 1). Then the Bargmann transformation ≤ ρ′ − 2,A deﬁned by (48) maps the space Wρ isomorphically onto E¯ρ , where A = 1/(2~). Proof. The transformation (48) is the composition of the following three operations: (1) 1 2 z¯ convolution by the function e− 2~ , (2) dilation by a factor of √2, and (3) multiplication 1 2 n/4 ~ z¯ by (π~)− e 2 . The ﬁrst operation is equivalent to the multiplication of ψˆ(σ) by ~ 2 2 σ 2 2 ce− , where σ = j σj and the precise value of the constant c is of no importance ~ 2 P ρ′,b n σ for the proof. Let ψˆ Wρ,a (R ) and ψˆ (σ)= ψˆ(σ)e− 2 . Then we have ∈ 1 ~ ′ (σ2 τ 2) a σ ρ+b τ ρ ψˆ (σ + iτ) ψˆ e− 2 − − | | | | . | 1 | ≤ k ka,b ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 15

Taking the inverse Laplace transform, setting z = x+iy, and using the Cauchy-Poincar´e theorem, we get

n/2 i(x iy)(σ+iτ) ψ1(¯z) (2π)− ψˆ a,b inf e − ψ1(σ + iτ) dσ | |≤ k k τ Z ≤ ~ ′ ~ xτ+ τ 2+b τ ρ yσ σ2 a′ σ ρ Ca′ ψˆ a,b inf e− 2 | | sup e − 2 − k k , ≤ k k τ σ

′ where a < a/nρ/2. Because ρ 2, we have τ 2 n + τ ρ , and the inﬁmum over τ ′ ′ ≥ ≤ j | j| can hence be estimated in a manner similar to that usedP in proving Theorem 4. Setting B′ = b + ~/2, we obtain

~ ~ ~ 2 ρ′ n ρ′ n ρ inf xτ + τ + b τ + inf  xτ + B′ τj  B x , τ − 2 | | ≤ 2 τ − | | ≤ 2 − | | Xj   ρ ρ′ where (B′ρ′) (Bρ) = 1. We now show that there exists a positive constant ǫ such that ~ 2 ρ 1 2 ρ yσ σ a′ σ y ǫ y for σ ~/2. (52) 2 2~ − − k k ≤ − k k k k≥ p After a suitable rescaling of the variables, (52) takes the form 2 ρ ρ (y σ) ǫ′ y a′′ σ , (53) − ≥ k k − k k ρ/2 ρ/2 where ǫ′ = (2~) ǫ, a′′ = (2/~) a′, and σ 1. Let us consider the ray y = tσ, t 0 in R2n. Because ρ 2 and σ 1, thek k inequality ≥ (53) holds on this ray if ≥ ≤ k k≥ 2 ρ (t 1) ǫ′t a′′. − ≥ − 2 2 This condition in turn is satisﬁed for all t if (t 1) ǫ′t a′′, or equivalently, ǫ′ − ≥ − 2 ρ ≤ a /(1 + a ). We note that for all σ in the ball σ 1, we have (y σ) ǫ y C ′ ′′ ′′ k k≤ − ≥ ′k k − ǫ with some constant Cǫ′ > 0. Thus, there exists a suﬃciently small positive number r(a, b, ~) such that

1 2 ρ ρ ~ y r x r y ψ (x iy) C′ ′ ψˆ e 2 − k k − k k . (54) | 1 − |≤ a k ka,b 1 2 ~ z¯ Therefore, the function ϕ(¯z)= e 2 ψ1(√2z ¯) satisfies (50) with A = 1/(2~) and with a preexponential factor proportional to ψˆ a,b. We conclude that the Bargmann operator ρ′ 2,1/2~ k k B maps Wρ continuously to E¯ρ . ~ 1 2 2,1/2 ~ z¯ Conversely, let ϕ E¯ρ and ϕ r < . Then the function ϕ1(¯z)= e− 4 ϕ(¯z/√2) satisfies ∈ k k ∞ 1 2 y r′ x ρ r′ y ρ ϕ (¯z) ϕ e 2~ − k k − k k , | 1 | ≤ k kr 1+ρ/2 where r′ = r/2 . Hence its Fourier-Laplace transform satisfies the estimate

n/2 i(x iy)(σ+iτ) ϕˆ1(σ + iτ) (2π)− inf e− − ϕ1(x iy) dx | |≤ y Z − ≤ 1 2 ′ ρ ρ yσ+ ~ y r y x τ b x Cb ϕ r inf e− 2 − k k sup ek k·k k− k k , (55) ≤ k k y x with b < r′. Substituting y = ~σ, we obtain ~ 1 2 ρ 2 ρ ρ inf yσ + y r′ y σ r′~ σ . y − 2~ − k k ≤− 2 − k k 16 M. A. SOLOVIEV

ρ′ Taking the supremum over x gives the conjugate convex function b′ τ . Therefore, it follows from (55) that k k

~ 2 ′ (σ+iτ) a σ ρ+b′′ τ ρ e 2 ϕˆ (σ + iτ) C ϕ e− | | | | , 1 ≤ k kr

ρ ρ′/2 1 ρ′ where a = r′~ and b′′ = b′n . We conclude that B− ϕ Wρ and the inverse 2,1/2~ ρ′ ∈ Bargmann transformation is continuous from E¯ρ to Wρ . The theorem is proved.

Corollary 1. For any ρ satisfying 1 < ρ 2, the Wick correspondence (49) has ≤ ρ n a unique extension to a continuous algebra homomorphism from ( (R ), ⋆Wick) to 2 E ¯2,A (F ), where A = 1/(2~). AEρ

9. Concluding remarks In this paper, we consider only analytic symbols of operators, but the above method of extending the star products and their representations by continuity, starting from a ρ ρ′ suitable auxiliary space like W or Wρ , has a wider ﬁeld of application. In particular, this approach allows us to extend the Weyl symbol calculus beyond the traditional framework of tempered distributions, see [20], [21]. Such a generalization is also desir- able for the anti-Wick correspondence because not all bounded operators have anti-Wick symbols among the usual functions [18]. ρ′ If ρ> 2, then the space Wρ is not an algebra under the Wick star product. However it is a topological algebra with respect to the Weyl-Moyal product ⋆~. It follows from a 1/ρ theorem of [17] that the elements of ρ are two-sided multipliers of the algebra (S , ⋆~) E 1/ρ ρ′ ρ which coincides with the algebra (Wρ , ⋆~) for ρ > 1. This implies that for all f ρ′ ρ′ ∈ E and ϕ Wρ , the products f ⋆~ ϕ and ϕ ⋆~ f belong to Wρ and are continuous in f and ϕ.∈ Furthermore, the following associativity relations hold:

(f ⋆~ ϕ)⋆~ ψ = f ⋆~ (ϕ⋆~ ψ), (ϕ⋆~ f)⋆~ ψ = ϕ⋆~ (f ⋆~ ψ), (ϕ⋆~ ψ)⋆~ f = ϕ⋆~ (ψ ⋆~ f). As shown in [21], the Weyl correspondence defines an isomorphism between the left ρ′ 2 multiplier algebra of (Wρ , ⋆~) and the operator algebra ρ′ (L ). Therefore, for the AWρ case of Weyl-Moyal product, Theorem 6 can be derived as a consequence of these results. The direct proof given here is more illuminating and applies to any star product of the form (2). Finally, we note that along with the spaces ρ, it is natural to consider the spaces of entire functions of the same order and finite type,E i.e., satisfying the condition a z ρ f(z) Ce | | , | |≤ where a and C are constants depending on f. We let Eρ denote this space and endow it with the inductive limit topology defined by the family of Banach spaces Eρ,a with a z ρ the norms f a = supz f(z) e− | | . It is easy to see that an analogue of Theorem 2 holds for Ekρ,k but only| if the| strong inequality ρ < 2 holds. It is essential that if 1/ρ′ + 1/ρ = 1 and ρ 2, then the series defining the star product f ⋆ϑ g converges ′ ≤ ′ absolutely in Eρ for all f ρ and g Eρ . Moreover, it can be proved that the ρ ∈ E ∈ ρ′ ρ′ algebra ( , ⋆ϑ) acts continuously on the space E and on its subspace Wρ by the left E ρ′ ρ′ and right ⋆ϑ-multiplication. In other words, the function spaces E and Wρ (as well as the analogous spaces associated with cone-shaped regions) are topological bimodules over the algebra ( ρ, ⋆ ) for any ϑ. E ϑ ALGEBRAS WITH CONVERGENT STAR PRODUCTS AND THEIR REPRESENTATIONS 17

Acknowledgements. This paper was supported in part by the the Russian Foundation for Basic Research (Grant No. 12-01-00865) References [1] H. Omori, Y. Maeda, N. Miyazaki, and A. Yoshioka, “Deformation quantization of Fréchet-Poisson algebras – Convergence of the Moyal product,” in: Conference Moshe Flato 1999. Quantization, Deformation, and Symmetries, edited by G. Dito, D. Sternheimer, Math. Physics Studies, no. 22, (Dordrecht, Kluwer, 2000), Vol. 2, pp. 233-245. [2] H. Omori, Y. Maeda, N. Miyazaki, and A. Yoshioka, “Non-formal deformation quantization of Fréchet-Poisson algebras: The Heisenberg Lie algebra case,” Contemp. Math. 434 (2007) 99-124. [3] S. Beiser, H. Römer, and S. Waldmann, “Convergence of the Wick product,” Commun. Math. Phys. 272 (2007) 25-52 [arXiv:math.QA/0506605]. [4] S. Beiser and S. Waldmann, “Fréchet algebraic deformation quantization of the Poincarédisk,” e-print arXiv:1108.2004. [5] O. W. Greenberg, “Failure of microcausality in quantum field theory on noncommutative space- time,” Phys.Rev. D 73 (2006) 045014 [arXiv:hep-th/0508057]. [6] M. A. Soloviev, “Noncommutativity and θ-locality,” J. Phys A: Math. Theor. 40 (2007) 14593- 14604 [arXiv:0708.1151]. [7] M. Chaichian, M. N. Mnatsakanova, A. Tureanu, and Yu. A. Vernov, “Test function space in noncommutative quantum field theory,” JHEP 0809 (2008) 125 [arXiv:0706.1712]. [8] M. A. Soloviev, “Failure of microcausality in noncommutative field theories,” Phys. Rev. D 77 (2008) 125013 [arXiv:0802.0997] [9] S. Galluccio, F. Lizzi, and P. Vitale, ”Twisted noncommutative field theory with the Wick-Voros and Moyal products,” Phys. Rev. D 78 (2008) 085007 [arXiv:0810.2095]. [10] A. P. Balachandran and M. Martone, ”Twisted quantum fields on Moyal and Wick-Voros planes are inequivalent,” Mod. Phys. Lett. A 24 (2009) 1721-1730 [arXiv:0902.1247]. [11] A. P. Balachandran, A. Ibort, G. Marmo, and M. Martone, “Inequivalence of QFTs on noncommutative spacetimes: Moyal versus Wick-Voros,” Phys. Rev. D 81 (2010) 085017 [arXiv:0910.4779]. [12] M. A. Soloviev, “Noncommutative deformations of quantum field theories, locality, and causality,” Theor. Math. Phys. 163 (2010) 741-752 [arXiv:1012.3536]. [13] M. Blaszak and Z. Domański, “Phase space quantum mechanics,” Ann. of Phys. 327 (2012) 167- 211 [arXiv:1009.0150]. [14] H. Omori, Y. Maeda, N. Miyazaki, and A. Yoshioka, “Deformation expression for elements of algebras (II) (Weyl algebra of 2m-generators),” e-print arXiv:1105.1218. [15] I. M. Gelfand and G. E. Shilov, Generalized Functions, Vol. 3 (Academic, New York 1968). [16] J. M. Gracia-Bondia and J. C. Várilly, “Algebras of distributions suitable for phase-space quantum mechanics. I.” J. Math. Phys. 29 (1988) 869-879. [17] M. A. Soloviev, “Twisted convolution and Moyal star product of generalized functions,” Theor. Math. Phys. 172 (2012) 885-900 [arXiv:1208.1838]. [18] G. B. Folland, Harmonic Analysis in Phase Space, Annals of Math. Studies, Vol. 122 (Princeton Univ. Press, Princeton, 1989). [19] F. A. Berezin and M. A. Shubin, Schrödinger Equation (Dordrecht, Kluwer, 1991). [20] M. A. Soloviev, “Moyal multiplier algebras of the test function spaces of type S,” J. Math. Phys. 52 (2011) 063502 [arXiv:1012.0669]. [21] M. A. Soloviev, “Generalized Weyl correspondence and Moyal multiplier algebras,” Theor. Math. Phys. 173 (2012) 1359-1376.

I. E. Tamm Department of Theoretical Physics, Lebedev Institute of Physics, Leninsky Prospect 53, 119991 Moscow, Russia E-mail address: [email protected]