Skip to main content
Log in

On the uniqueness of the position operator for relativistic elementary systems

  • Published:
Il Nuovo Cimento (1955-1965)

Summary

The regularity condition first introduced by Newton and Wigner is here re-examined in some detail for the case of a relativistic elementary system [m≠0, s]. Three different types of normalizable sequences ψn approaching some localized state ψa are considered, namely: A) {ie413-01} in the topology of {ie413-02} uniformly on any compact set; and C) ψ nare characteristic functions inq-space with compact supports such that suppq ψn⊃ suppg ψn+1 ) a, μ(suppq ψn)→ 0. It is then proved that even the ∞-order regularity condition is satisfied by all possible position operatorsQ if A) sequences are allowed, while the restriction toB) or C) sequences rendersQ unique as soon as the 2nd order regularity condition is imposed. Finally a different but related abstract characterization is provided for the standardQ (0)as being that (unique) position operator for whichQ 2is essentially selfadioint when restricted to the domain {ie413-03}

Riassunto

Nel caso di un sistema elementare relativistico (m≠O, s) si riesamina in dettaglio la condizione di regolarità, introdotta per primi da Newton e Wigner. Si prendono in considerazione tre diversi tipi di sequenze normalizzabili ψn, che approssimano uno stato looalizzato ψa, cioè: A, {ie422-01} nella topologia di {ie422-02} uniformemente su ogni gruppo compatto ; eC, ψ nsono funzioni caratteristiclie nello spazioq con supporti compatti, tali che {ie422-03}. Si dimostra poi che anche la condizione di regolarità di ordine ∞ è soddisfatta da tutti i possibili operatori di posizioneQ se si ammettono le sequenzeA, mentre la restrizione alle sequenzeB o C rendeQ unico non appena si impone la condizione di regolarità di secondo ordine. Inline si dà una caratterizzazione astratta diversa, collegata alla précédente, per ilQ (0)standard, cioè per quell’(unico) operatore di posizione per cui Q2 è essenzialmente autoaggiunto quando è ristretto al dominio {ie422-04}

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

eferences

  1. T. D. Newton andE. P. Wigner:Rev. Mod. Phys.,21, 400 (1949).

    Article  ADS  MATH  Google Scholar 

  2. A. S. Wightman:Rev. Mod. Phys.,34, 845 (1962).

    Article  MathSciNet  ADS  Google Scholar 

  3. We use the same notation as inW. Weidlich andA. K. Mitra:Nuovo Cimento,30, 385 (1963). See this reference for further details.

    Article  MathSciNet  MATH  Google Scholar 

  4. L. Schwartz:Théorie des Distributions, I, II (Paris);N. Dunford andJ. T. Schwartz:Linear Operators, I, II (New York).

  5. By position operator we will mean any operatorQ satisfying the Newton-Wigner postulates a), b), e), given in (1).

  6. It should be noted that the setR is here defined in afixed realization of [m,s]; since continuity of ψa is not an intrinsic (i.e., realization-independent) property,R will be defined by unitary equivalence in other realizations. An intrinsic characterization ofSi will be given, however, in Sect. 2.

  7. See (1), p. 404.

  8. This contradicts BC1 for s>0.

  9. In (3) the authors claim thatQ is unique independently of any EC. This result is in contradiction with those in (1,2) and with the present paper.

  10. This is a consequence of the following statement: if 421-01 and if all the derivatives ofT of order [n/2] + l belong toL1, thenT is a continuous function. (See L. Schwartz: loc. cit., Livre II, p. 37, 38, 44).

  11. (17) Note that in the 3-dimensional space every function in the domain of the Laplacian operator is equivalent to a continuous function.

  12. It suffices to show this for the Laplacian operator A inB3, i.e., we must find somefeL2(B3) such that [(à + Z)g, /] = 0 for some complexZ (with ImZ^O) and allgel), D being a subdomain of I>A consisting of functions vanishing at some point, say j(X0) = 0. By Fourier-transforming the problem, one immediately realizes that^X) ^j^TT*exp [-^x-xo)]d3C is the desired function.

Download references

Author information

Authors and Affiliations

Authors

Additional information

On leave of absence from the Junta de Energia Nuclear, Madrid.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Galindo, A. On the uniqueness of the position operator for relativistic elementary systems. Nuovo Cim 37, 413–422 (1965). https://doi.org/10.1007/BF02749843

Download citation

  • Received:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF02749843

Navigation