You are on page 1of 5

24 May 2001

Physics Letters B 508 (2001) 211–215


www.elsevier.nl/locate/npe

Possible vacuum-energy releasing


She-Sheng Xue
ICRA, INFN and Physics Department, University of Rome “La Sapienza”, 00185 Rome, Italy
Received 18 January 2001; accepted 10 April 2001
Editor: R. Gatto

Abstract

We study a variation of the vacuum-energy when external classical fields are applied upon the vacuum. Taking a constant
magnetic field as an example, we discuss why and how the vacuum-energy should be released in the context of quantum field
theories and we give theoretical computations showing how much vacuum-energy can be released.  2001 Published by Elsevier
Science B.V.
PACS: 12.20.Ds; 12.20.Fv

The vacuum has a very rich physical content in |E| ∼ V Λ4p , which is a tremendous amount of ener-
the context of relativistic quantum field theories. It gies for a macroscopic volume V . According to quan-
consists of extremely large number of virtual particles tum field theories for free and massive particles, the
and anti-particles. The quantum fluctuations of the energy-spectrum of virtual particles in the vacuum is
vacuum are creations and annihilations of these virtual given by
particles and anti-particles in all possible energy- 
range. As a consequence of the quantum fluctuations, (k) = ± (ckx )2 + (cky )2 + (ckz )2 + m2 c4 , (2)
the vacuum-energy in a volume V of the three-
where m is the mass of particles, the sign “+” is for
dimensional space is given by
electromagnetic fields and “−” for fermionic fields.

E = sV (k), (1) In the case of electromagnetic fields (m = 0), the
k virtual photon and its anti-particle are self-conjugated
and all virtual photons can be accommodated in the
where the (k) is the energy-spectrum of virtual parti-
same energy-momentum state. The positive vacuum-
cles, the summation is over all possible momentum-
energy (1) is due to the quantum fluctuations of these
states of quantum-field fluctuations and the factor s
virtual photons in all possible energy-range. In the
is 1/2 for electromagnetic fields and 1 for fermionic
case of fermionic fields, a pair of virtual fermion and
fields. The vacuum-energy (1) is related to the tran-
anti-fermion occupy a energy-momentum state and
sition amplitude 0|0 from the vacuum to the vac-
all negative energy states of quantum fermion fields
uum, which is represented by close loops for virtual
down to the negative energy-level of the Planck scale
particles in Feynman diagrams. Up to the fundamen-
(−Λp ) have been fully filled by the pairs of virtual
tal Planck scale Λp ∼ 1019 GeV, the vacuum-energy
fermions and anti-fermions. The virtual fermion in
the negative energy state travels backward in time
E-mail address: xue@icra.it (S.-S. Xue). indicating a virtual anti-fermion in the positive energy
0370-2693/01/$ – see front matter  2001 Published by Elsevier Science B.V.
PII: S 0 3 7 0 - 2 6 9 3 ( 0 1 ) 0 0 4 8 2 - 8
212 S.-S. Xue / Physics Letters B 508 (2001) 211–215

state travels forward in time. The negative vacuum- energy computed by the energy-spectrum (2), an ex-
energy (1) is due to the quantum fluctuations of the ternal force must be introduced to against a repulsive
creations (annihilations) of virtual fermions and anti- force and store the work done by the external force
fermions by (to) virtual photons in all possible energy- into the vacuum-energy.
range. The pairs of virtual fermions and anti-fermions In this Letter, instead of modifying the energy-
are created and annihilated in the time scale ∼ h̄/mc2 spectrum (2) of electromagnetic fields by boundary
and at the distance scale ∼ h̄/mc. conditions, we attempt to study the variation of the
In the description of renormalizable and perturba- vacuum-energy (1) by modifying the negative energy-
tive quantum field theories, the vacuum state is a spectrum (2) for virtual fermions with external clas-
ground state consisting of virtual particles with the sical fields and to find any possible effects of the
energy-spectrum (2), all negative energy states are vacuum-energy releasing due to such a modification.
fully filled and real particles are excitation quanta We first take the case of a constant magnetic field
upon this ground state. This description is validated, as an example to discuss and illustrate the reasons
provided the amplitude of quantum fields and the for the vacuum-energy releasing. The negative energy-
strength of interactions between quantum fields are spectrum (2) for virtual fermions is not degenerate.
small upon such a vacuum state. In this description, The vacuum state is made by virtual fermions fully
the vacuum-energy (1) is dropped and set to be zero filling this negative energy-spectrum. We apply the
by the normal ordering of creation and annihilation external constant magnetic field H onto such a vac-
operators, due to the absolute value of the physical en- uum state along the z-axis. This constant magnetic
ergy only determined up to a constant and the quantum field is confined within a finite space of a volume V =
fluctuation of the vacuum impacting on real particles Lx · Ly · Lz . As well known as the Landau levels [4],
is treated by the renormalization of theories. The de- the negative energy-spectrum of virtual charged fermi-
scriptions of renormalizable and perturbative quantum ons is given by
field theories have been extremely successful. As an
example, the Lamb-shift [1] effect and electric charge

renormalization indeed exhibit the highly non-trivial (kz , n, σ ) = − (ckz )2 + m2 c4
structure of the quantum electromagnetic dynamics 1/2
(QED) and its vacuum (ground) state. + h̄c|e|H (2n + 1) − h̄ceH σ ,
However, as shown by the Casimir effect [2] that
was experimentally evidenced [3], the positive vacu- n = 0, 1, 2, 3, . . ., (3)
um-energy (1) of virtual photons is not just a triv-
ial constant, when the quantum fluctuations of virtual where e and σ = ±1 are fermion’s charge and he-
photons of the vacuum state are confined within a fi- licity. This negative energy-spectrum is degenerate
nite volume by boundary conditions. The Casimir ef- in the phase space of (kx , ky ) and the degeneracy is
fect can be physically understood as the following: |e|H S/(2π h̄c), where the area S = Lx · Ly .
the continuous energy-spectrum (2) of electromag- The negative energy-spectrum is changed from (2)
netic fields is modified by boundary conditions to be to (3) due to the external magnetic field H . If the
discrete one, the “new” vacuum-energy computed by vacuum-energy of the “new” vacuum state made by
the modified energy-spectrum in a given finite volume virtual fermions fully filling the negative energy-
is smaller than the “old” vacuum-energy computed by spectrum (3) is smaller than that of the “old” vacuum
the energy-spectrum (2) in the same volume. As a re- state made by virtual fermions fully filling the negative
sult, the vacuum has to quantum-mechanically fluctu- energy-spectrum (2), the “old” vacuum state must
ate from the “old” vacuum state to the “new” vacuum quantum-mechanically fluctuate to the “new” vacuum
state. This leads to an attractive force observed in the state. The difference of vacuum energies between two
Casimir effect. This is in fact a phenomenon of extract- vacuum states must be released. In order to verify
ing vacuum-energy from the vacuum. On the contrary, this and possibly observable effects, we are bound to
if the “new” vacuum-energy computed by the modi- compute the energetic difference between two vacuum
fied energy-spectrum is larger than the “old” vacuum- states corresponding to H = 0 and H = 0.
S.-S. Xue / Physics Letters B 508 (2001) 211–215 213

The total energies of the “old” vacuum state and The analytic continuation “z” has simply discarded the
“new” vacuum state are respectively given by, appropriate divergent terms and the continuation back
  to “z = 1” ( → 0) yields,
V   
Eo = −4 d ck (k), (4)  
(2π h̄c) 3  |e|2H 2 π(m2 c4 )2
En = V  − + . (10)
and 2 3(4π 2 h̄c) (2π h̄c)3
   
|e|H S Lz  As a result, the energetic difference of “new” and
En = − dkz (kz , n, σ ). (5)
2π h̄c 2π h̄ “old” vacuum states is,
n,σ  
 |e|2 H 2 V
Both vacuum energies are divergent up to the Planck E = En − Eo =  − . (11)
scale −V · Λ4p . The difference of vacuum energies be- 2 3(4π 2 h̄c)
tween “new” and “old” vacuum states can be com- We find that in Eq. (10) the term depending on the
puted by the approaches of the dimensional regulariza- fermion mass is completely canceled by Eq. (6),
tion [5] and ξ -function regularization [6]. In Eq. (4), as it should be. For  → 0, the gamma-function
analytically continuing the dimension of the momen- (−/2) = −(2/ + const.), where the constant is an
tum integration from 3 to 3 + , where  is a small uninteresting combination of π , γ (Euler constant),
complex parameter, we have etc. On the basis of the charge renormalization of the
    √
Vπ  2 4  QED, we renormalize the charge |er | = Z3 |e| by the
Eo = mc  − , (6) renormalization constant Z3 = (2/ + const.) and we
(2π h̄c)3 2
obtain,
where (x) is the gamma-function. Analogously, in
  e2 H 2 V α
Eq. (5), analytically continuing the dimension of the E = − Q2f r
= −8 H 2 V ,
2
momentum integration from 1 to 1 + , we have, f
3(4π h̄c) 3π
   
|e|H V  er2
En = −  − α= , (12)
4π 2 h̄2 c2 2 4π h̄c
 z
× m c + h̄c|e|H (2n + 1) − h̄ceH σ , (7)
2 4
where the fine structure constant α = 1/137 and
f (Qf ) = 8 for all charged fermions in the standard
n,σ 2

where z = 1 + /2. Summing over helicity states σ = model of particle physics. Eq. (12) is about one percent
±1, we obtain of the total energy deposited by the external magnetic
    field H .
|e|H V  In principle, since E (12) is negative, indicating
En = −  −
4π 2 h̄2 c2 2 that the energy of the “new” vacuum for H = 0 is
 z
× m c + 2h̄c|e|H n
2 4 smaller than that of the “old” vacuum for H = 0,
n  z the vacuum must release the energy (12) when the
+ m2 c4 + 2h̄c|e|H (n + 1) external magnetic field is applied upon it. Let us
   
2|e|2H 2 V  try to understand why and how this could physically
=−  −
4π 2 h̄c 2 occur. This can possibly occur for three reasons
× ξ(−z, q) + ξ(−z, q + 1) , that (i) in a finite volume V , the total number of
fermionic states in the vacua of negative energy-
m2 c 4 spectrum (2) and (3) is finite for the finite momentum-
q= , (8)
2h̄c|e|H cutoff at the Planck scale Λp and all these fermionic
where ξ(z, q)-function is given by [7], states of negative energy levels from −Λp to −mc2
 are fully filled; (ii) the negative energy-spectrum
1 B2 (q)
ξ(z, q) = , ξ(−1, q) = − , (2) is not degenerate, while the negative energy-
(n + q)z 2 spectrum (3) is degenerate, and the total numbers of
n=0
1 fermionic states of both cases are the same as indicated
B2 (q) = q 2 − q + . (9) in (i); (iii) on the basis of quantum fluctuations
6
214 S.-S. Xue / Physics Letters B 508 (2001) 211–215

toward the lowest energy-state and the Pauli principle, fluctuations of vacuum states should be the order of
when the external magnetic field is applied upon the h̄/me c2 ∼ 10−20 s. With this very short relaxation
vacuum, the vacuum reorganize itself by fully filling time, all causally-correlated vacuum states at differ-
all fermion states according to the degenerate negative ent points of the space–time can rapidly decay to the
energy-spectrum (3), instead of the non-degenerate lower energy states before the maximum value of the
one (2). As a consequence, the vacuum makes its total external magnetic field is reached. In this case, we
energy lower and releases energy. As an analogy, the should not expect to detect appropriate photon emis-
vacuum with the negative energy-spectrum (2) can be sions, corresponding to the total vacuum-energy re-
described as if a N -floors building, two rooms each leasing given in Eq. (12), only from the volume V
floor and all rooms occupied by guests; while the where the constant magnetic field is created, since
vacuum with the negative energy-spectrum (3) a M- photons can be spontaneously emitted from the whole
floors (M < N) building, 2N/M rooms each floor volume ∼(τ · c)3 where causally-correlated vacuum
and all rooms occupied by guests. The total numbers states are. If we can create the magnetic field in a short
of rooms of two buildings are the same. Due to an enough time, we should expect to detect photon emis-
external force, the N -floors building collapses to the sions and other phenomena discussed in the previous
M-floors building and the “potential energy” should paragraph. Based on these discussions, it seems very
be released. hard to experimentally detect such a vacuum-energy
If the vacuum-energy (12) is released, the phenom- releasing. We would expect a very sophisticated ex-
ena and effects that could occur are following. (i) The periment to verify the phenomena and effects of the
vacuum acts as a paramagnetic medium that effec- vacuum-energy releasing induced by the external mag-
tively screens the strength of the external magnetic netic field. This is important for the understanding of
field H to a smaller value H  < H for the total energy- the fermionic structure of the vacuum of relativistic
density being quantum field theories and any possibly prospective
 applications. 1
1 2 1 2 α 2  16 As a comparison, we consider the case of an exter-
H = H −8 H , H =H 1− α.
2 2 3π 3π nal electric field by a simple example. For a spheri-
(13) cal conductor with the radius R◦ and charge Q◦ < 0,
we have the electric potential U and field E, U =
(ii) Photons are spontaneously emitted analogously U◦ = Q◦ /R◦ and E = 0 for r  R◦ ; U = Q◦ /r and
to the spontaneous photon emission for electrons at E = Q◦ /r 2 for r > R◦ . We postulate the electric po-
high-energy levels decaying to low-energy levels in tential energy eU◦ > mc2 and the field E at R◦ is
the atomic physics. (iii) This strong vacuum-energy less than the critical value to create electron–positron
fluctuation could lead to the emission of neutrino and pairs. It is rather complicate to find the exact negative
anti-neutrino pairs from the vacuum, since they are energy-spectrum in such an attractive potential. To ob-
almost massless. tain a preliminary result, we may approximate this po-
In practice, on the other hand, how the phenomena tential by a spherical potential well U = U◦ , r  R◦ ;
of the vacuum-energy releasing can be realized and U = 0, r > R◦ . The first negative discrete energy-level
observed in reality. In the theoretical analysis given, E1 of an electron of charge e < 0 is given by [8]
we actually assume that the transition of the vacuum
states from the negative energy-spectrum (2) to the π 2 min 2 (π h̄c)2
E1 = − U  , U◦min = ,
negative energy-spectrum (3) is instantaneous. In re- 16 ◦ 8mc2R◦2
ality, there is a time scale τ for building up the  2
magnetic field to its maximum value. The maximum eU◦
2 = − 1 , (14)
volume (τ · c)3 , in which vacuum states at differ- U◦min
ent points of the space–time are causally-correlated,
can be much larger than the volume V in which
the constant magnetic field is created. In addition, 1 The right to use this idea (as an intelligent product) for any
we know that the relaxation time t of quantum applications is preserved by the author.
S.-S. Xue / Physics Letters B 508 (2001) 211–215 215

for small , where U◦min is the minimum depth of upon the vacuum, we illustrate this idea in very details
the potential well for appearing the first negative and give an explicit computation showing the vacuum-
discrete energy-level. For the macroscopic value of energy releasing is about one percent of the total en-
the radius R◦  h̄/(me c), U◦min and E1 are extremely ergy stored by the external magnetic field. Based on
small. Based on this approximate negative energy- general discussions on time scales of quantum fluc-
spectrum, we can estimate the difference between the tuations and creating a macroscopic magnetic field, it
total vacuum energies En with and Eo without the seems difficult to detect such vacuum-energy releasing
presence of the external electric potential U◦ , in practice.
E = En − Eo  E1 < 0. (15)
This simply shows that the negative energy-spectrum References
of the vacuum is shifted downward from −mc2 to
−mc2 + E1 . The possible vacuum-energy releasing [1] W.E. Lamb, R.C. Retherford, Phys. Rev. 72 (1947) 241.
|E1 | in this case is extremely small for the radius [2] H.B.G. Casimir, Proc. Kon. Nederl. Akad. Wetensch. B 51
R◦  h̄/(mc). The result of the similar idea applying (1948) 793;
to the case of an external gravitational field will be See also, M. Fierz, Helv. Phys. Acta 33 (1960) 855.
presented in another paper [9]. [3] M.J. Sparnnaay, Physica 24 (1958) 751;
S.K. Lamoreaux, Phys. Rev. Lett. 78 (1957) 5;
In summary, we propose an idea of releasing vacu- U. Mohideen, A. Roy, Phys. Rev. Lett. 81 (1998) 4549.
um-energy from the vacuum by introducing external [4] D. Landau, M. Lifsits, Theory of Relativistic Quantum Mechan-
classical-fields. This idea is originated from (i) the ics, 1975, p. 357.
fermionic structure of the vacuum owing to relativis- [5] G. ’t Hooft, M. Veltman, Nucl. Phys. B 44 (1972) 189.
tic quantum field theories and the Pauli principle; (ii) [6] J.S. Dowker, R. Banach, J. Phys. A 11 (1978) 2255.
external classical-fields possibly modifying the fermi- [7] I.S. Gradshteyn, I.M. Ryzhik, Table of Integrals, Series, and
Products, Academic Press, New York, 1965, (9.521), (9.531)
onic structure of the vacuum; (iii) the quantum fluc- and (9.627).
tuations of virtual particles in the vacuum must lead [8] D. Landau, M. Lifsits, Quantum Mechanics, 1975, p. 143.
the vacuum to the lowest energy-state. Taking a par- [9] S.-S. Xue, in preparation.
ticular example of the constant magnetic field applied

You might also like