VOLUME 51, NUMBER 14 PHYSICAL REVIEW LETTERS 3 OCTOBER 1983

N u c l e i : A Superfhiid C o n d e n s a t e of a P a r t i c l e s ? A S t u d y w i t h i n t h e I n t e r a c t i n g - B o s o n M o d e l
Y. K. Gambhir ( a > and P . Ring
Physik-Department, Technische Universitat Munchen, D-8046 Garching, West Germany
and
P . Schuck
Institut des Sciences Nucleaires, F-38042 Grenoble, France
(Received 11 July 1983)
The authors have studied the question of whether pairs of neutrons and pairs of protons
of the usual superfluid phases form a bound state to give rise to a superfluid condensate
of "a particles." They indeed find indications for this to be the case from a BCS-like
study for bosons using the proton-neutron interacting-boson model as well as from an
even-odd effect in the number of pairs using experimental binding energies.
PACS numbers: 21.65.+f, 21.60.Dr, 21.60.Gx
The question of a - p a r t i c l e c o r r e l a t i o n s in n u - E i c h l e r and Y a m a m u r a 2 exhibits many f e a t u r e s
c l e i and w h e t h e r nuclei a r e a superfluid c o n d e n - which we will a l s o find in our study. O u r r e s u l t s
s a t e of a p a r t i c l e s r a t h e r than an o r d i n a r y can, of c o u r s e , only be v e r y qualitative but on
f e r m i o n - p a i r condensate h a s long 1 " 3 b e e n r a i s e d this level we find s o m e intriguing a g r e e m e n t with
(we h e r e p r e f e r to u s e the t e r m a p a r t i c l e in a e x p e r i m e n t ; this c o n c e r n s , for i n s t a n c e , the
b r o a d s e n s e to indicate any kind of bound s t a t e quite pronounced even-odd effect in the n u m b e r
between two p r o t o n s and two n e u t r o n s r a t h e r of p a i r s and its dependence on the n e u t r o n e x c e s s
than " q u a r t e t " o r " q u a d r u p e l " ) . F o r i n s t a n c e which we found f r o m the study of e x p e r i m e n t a l
Danos and Gillet and A r i m a and Gillet, 1 E i c h l e r binding e n e r g i e s and which can be c o n s i d e r e d a s
and Y a m a m u r a , 2 and K a m i m u r a , M a r u m o r i , and the analog of the even-odd effect of o r d i n a r y n u -
Takada 3 have investigated this p r o b l e m in quite clear pairing.
s o m e detail m o r e than a decade ago. M o r e r e - Our s t a r t i n g point i s the phenomenological
cently this old question s e e m s to have r e c e i v e d Hamiltonian of the n e u t r o n - p r o t o n i n t e r a c t i n g -
r e n e w e d attention, 4 * 5 p a r t l y b e c a u s e of the s u c - b o s o n model in its s i m p l e s t v e r s i o n s i n c e we
c e s s of the p r o t o n - n e u t r o n i n t e r a c t i n g - b o s o n h e r e want to make only a qualitative study (we
model (IBM2). 6 But a l s o D u s s e l , Lotta, and u s e s t a n d a r d notation 6 ):
P e r a z z o 5 pointed out not long ago that c o n d e n s a -
tion of a p a r t i c l e s might indeed b e plausible H =Z/ pessp % + €
a d T
p d
p - KQirQv (1)

f r o m a s y s t e m a t i c investigation of e x p e r i m e n t a l
a-particle Q values.
As a m a t t e r of fact the question w h e t h e r p a i r s
of f e r m i o n s condense a s such or a s b i p a i r s i s an
i m p o r t a n t question not only in a t o m i c nuclei,
but h a s its r e l e v a n c e in s e v e r a l other d o m a i n s
of p h y s i c s a s well. Indeed our investigation has
b e e n i n s p i r e d by a r e c e n t w o r k of N o z i e r e s and
S a i n t - J a m e s 7 w h e r e the condensation of excitons
v e r s u s biexcitons in a s e m i c o n d u c t o r has b e e n
studied in an i n t e r a c t i n g - b o s o n model with u s e of
a BCS f o r m a l i s m for b o s o n s .
We h e r e e s s e n t i a l l y follow t h e i r i d e a s , a p p l y -
ing t h e m to IBM2. T h i s can, of c o u r s e , only be
a quite c r u d e model for the question of a - p a r t i c l e
condensation s i n c e the underlying f e r m i o n s t r u c -
t u r e of the bosons c e r t a i n l y p l a y s an i m p o r t a n t
r o l e . A completely f e r m i o n i c t r e a t m e n t of the
p r o b l e m s e e m s , however, quite difficult though
s o m e e a r l y a t t e m p t s in this d i r e c t i o n exist. 2 , 3 * 8
F o r i n s t a n c e , the exactly solvable model by
with
Qpu^dp^Sp+Sp^dpn + xidpUp]^ ,
/J
w h e r e rf/i = (-) ^. / [ I and p s t a n d s for p r o t o n (n) o r
neutron (v). The o p e r a t o r s sp and dp a r e p u r e
boson operators approximating proton or neutron
p a i r s coupled to angular m o m e n t u m 0 and 2, r e -
spectively, thus r e t a i n i n g the m o s t i m p o r t a n t d e -
g r e e s of f r e e d o m of n u c l e a r d y n a m i c s . The p a -
r a m e t e r s e n t e r i n g (1) have b e e n adjusted f r o m
e x p e r i m e n t and it i s known that one obtains a
quite good d e s c r i p t i o n of the s y s t e m a t i c s of the
low-lying p a r t of the s p e c t r a of e v e n - e v e n n u -
clei. 6
The Hamiltonian (1) exhibits a quite s t r o n g
i n t e r a c t i o n between neutron and p r o t o n p a i r s and
it i s thus p r e d i s p o s e d for the study of a c o r r e l a -
t i o n s . The question whether (1) gives r i s e to a
condensation of a p a r t i c l e s can be i n v e s t i g a t e d
m o s t conveniently by g e n e r a l i z i n g the BCS f o r m a l -
i s m for f e r m i o n s (formation of f e r m i o n p a i r s ) to
bosons (formation of boson p a i r s ) (for a study of

© 1983 The American Physical Society 1235

VOLUME 51, NUMBER 14 PHYSICAL REVIEW LETTERS 3 OCTOBER 1983

this generalization see, e.g., Nozigres and Saint-James, 7 Schuck and Ethofer, 9 and Rig and Schuck10)o
We thus introduce the following Bogoliubov transformation among the boson operators s p a n d dp:
t 1 t =4
UsS* ~VSSV+X, 6
U
= // dU
va d V\l t -^S*+^i 'i/p
!
'd dv\l """ Vd ^7T|i> (2)
1TjLl ^d 1TJJ

with ui2-vi2 = 1; i = s, d. In contrast to the fermion case there can appear constant terms such as xv
and xv, the significance of which becomes clear if we write down the vacuum which corresponds to the
quasiboson operators (2)10:
|$> a e x p { * f f s / + * 1 / s I , T + c 4 t s l r t s I , t + cd[d17'tdv't]o}\0). (3)
Here 10) stands for the total vacuum and the new
parameters are related to the ones in Eqs. (2) as final expressions for our problem:
follows:
1 €,-A . .
= 5l/2vd/ud, (4) (8)
Zp = -x /u ;
- P - -~ p p / s
cs=vs/us;
The first two terms in the exponent of (3) c o r r e -
spond to the condensate of single bosons (of the £ i = [ ( € , - A)2 - A , 2 ] 1 / 2 ; UiV^-^Ai/Et, (9)
usual BCS type) which is taken care of by the con- As=-^5Ad/isd, (10a)
stants x TT and xv in (2) and the last two terms in
the exponent correspond to the boson-pair con-
densate. The parameter X is determined as usual from
We now look for the minimum of the energy the particle number condition and ju, =ES, We
with respect to the parameters of the transforma- see that Eqs. (8)-(10) resemble ordinary BCS
tion (2) under the constraints that the total num- equations for fermions with only a number of
ber of particles as well as the neutron excess signs reversed because of the boson structure of
has a definite value: our problem. For instance the quasiparticle en-
6(<p\H -X» - (i{nv -nv)\ <p) = 0 . (5) ergies contain now a difference of two positive
numbers under the square root where for fer-
The calculus is standard and leads to the follow- mions we have a sum. In our numerical study
ing set of coupled equations: however, it turns, out that the particle number
2uivi{ei-\)+ Ai{ui2+vi2)=0, i=s,d, (6a) condition always prevents the quasiparticle ener-
gies from becoming imaginary.
with One of the arguments for ordinary proton or
A 5 = 5nudvd, Ad=K{usvs + \2udvd +3> ^ J , (6b) neutron pairing in nuclei which is usually put for-
ward is the even-odd effect. This means that it
and usually costs more energy to take out a proton or
a neutron from an even-even nucleus than from
an odd one because of the extra binding of the
y v = -usxv-vsxVf (6c)
pairs. This gives rise to the very well known
sawtooth behavior for the Q values of neutrons
y v = -usxv-vaXv. (6d) (or protons) as a function of neutron (or proton)
number. In analogy to this we should find in the
Equations (6c) and (6d) have, besides the trivial case of a -particle superfluidity an even-odd ef-
solution yv=:y1T = 0, a nontrivial one which for the fect as a function of the number of pairs. This
case where we have neutron excess is given by implies that it should be more difficult to take
a neutron or proton pair out of a nucleus which
\l/2
yv = -usx; x=xv =(nv (i.e., in practice whose open-shell configuration)
(7) consists only of a particles than from a nucleus
y* = -vsx; x^O. which consists of a particles plus one odd pair.
In order to investigate this we took as our inert
In this case we have only neutrons in the single- core the doubly magic nucleus ^ S n ^ and filled
boson condensate of (3) and all protons are bound in eight a particles which leads to ^ D y ^ . We
in a particles. From (6a) and (6b) we obtain the then added an odd neutron or proton pair leading

1236
VOLUME 51, NUMBER 14 PHYSICAL REVIEW LETTERS 3 OCTOBER 1983

to looses a n d 198Er68> respectively. From their 16

! 1 1 1 1 1 1 1 i i i
averaged (in order to smooth out eventual isotop-
Er,7
ic effects) binding energies 1 1 we subtracted that
164
of Dy which, e.g., gives us the point at n = 17
(MeV)

U !
, / V 'ft "A * ^,
* "fi
6
Hf
-
-7 Y/W% 'V ,8°w «K
of the full line in Fig. 1. We then subtracted s
from the binding energy of J^Er 68 (nine a's) the
averaged binding energies of ^ E r ^ (minus one
-i5oDy W
12 - A
V
y A
A A Av Y
? '«Bp.>
y V\ /A \v/v * A ,96 pb
V \VA\ AW A 200PO
neutron pair) and }^Dy66 (minus one proton pair) /- Y y V ^ -AAAA \/V V H
Y v \' \ w /A\ 20A.4R"
which yields the point at n = 18 on the full line of
Fig. 1. This procedure was continued up to the 10
addition of eighteen a particles to the ^ S n ^ i i i i i i i i 1 1 1
16 18 20 22 2U 26 28 30 32 34 36 ,
core leading to the pronounced sawtooth behavior
FIG. 1. Even-odd staggering as a function of boson
as a function of the number of pairs as shown in number. The topmost full line corresponds to the chain
Fig. 1, corroborating thus the idea of a -particle beginning with 164Dy (as described in the text) and the
superfluidity. lower full line to the chain of 150Dy as obtained from the
Before discussing the significance of the other mass tables. The broken line corresponds to the p r e s -
curves in Fig. 1 let us return to our theory and ent calculation.
investigate how from there such an even-odd ef-
fect finds its natural explanation. We calculate
the ground-state energies for even and odd boson numbers using 132Sn as an inert core:
En°=(<f>\H\$) = 2esvs2 + 10edvd2-5K[2usvsudvd+X2(udvd)2\, (11a)
t 0 2 2
E„+1°=(#|o^ap |$)=^n -»-€5(^ +z;s )+10/c(^t;s)(^i;J, (lib)
and determine the parameter vd in both cases by
1
minimizing the energy. (vs is eliminated by the ferences considered here for the determination
constraint on the particle number.) In order to of certain parameters in the IBM2. Although the
take account of the A dependence of es and ed effect of deformation may be relatively small for
(which is certainly present over such a long energy differences considered here, one definite-
mass range as shown in Fig. 1) we made the fol- ly should allow for a deformed trial function (3).
lowing Ansatz: Nevertheless our model study seems to indicate
that protons and neutrons in doubly open nuclei,
€s,,=£s/[l32/(132+^)]l/3,
instead of forming two superfluids of proton and
where es° and g are adjustable parameters found neutron pairs, rather agglomerate to form a
to be 13.7 MeV and 0.15, respectively, in the particles which condense to a superfluid phase.
present analysis. The difference e = ed° - es° has Before drawing definite conclusions we should,
been taken to be 0.2 MeV, a value compatible however, keep several points in mind. The ef-
with the IBM2 parameters. For K we took 0.5 fects we are after are genuine four-fermion cor-
MeV and x was put equal to zero because we relations which therefore should not already be
found—within reasonable ranges—very little de- contained in a fermion wave function of the mean-
pendence of our results on this parameter and x field type. It is, however, clear that the asym-
= 0 can be considered as its mean value. With metry energy contained in the usual Hartree-
these parameters we then calculated from (8)- Fock-Bogoliubov theory gives rise, at least p a r -
(11) the broken curve of Fig. 1 which shows semi- tially, to the effects that we are here discussing
quantitative agreement with the corresponding in our interacting-boson model. How much of the
experimental data. effect is due to genuine four-fermion correla-
One certainly cannot expect more from our tions can in the end probably only be decided
theory and the present form of the IBM2 Hamil- from a true four-fermion calculation.
tonian where the internal structure of the bosons That the bosonic description might not be en-
is entirely neglected; one also might think that tirely devoid of exhibiting true four-body c o r r e -
for a more realistic investigation a more com- lations could, however, be concluded from the
plete IBM2 Hamiltonian including for instance following observation: Suppose that the nucleus
the Major ana force and other bosons should be Z=N= 50 existed; our full sawtooth line of Fig. 1
taken. 8 One then could even take the mass dif- already discussed above would consist then of n/

1237
VOLUME 51, N U M B E R 14 PHYSICAL REVIEW
2 a particles and sixteen unpaired neutron bosons.
Had we unpaired fermions instead of unpaired
bosons, the fermion-pair superconductivity would
be strongly suppressed because of the blocking
effect, as is very well known.10 On the contrary
in our case of boson-pair condensation the super-
conductivity effect is enhanced in the presence of
unpaired bosons as is, e.g., easily seen from Eq.
(10b) where the interaction strength K becomes
multiplied by a factor of 1 +x2 where x2 is the
number of unpaired bosons. This effect is also
verified experimentally and demonstrated in Fig.
1 where we show besides the sawtooth line for
sixteen neutron bosons plus n/2 a ' s also another
sawtooth line where we successively diminished
the number of unpaired bosons. We see that r e -
ducing the number of unpaired bosons from six-
teen to nine reduces the amplitude of the even-
odd staggering by about a factor of 2, a quite
nice demonstration of the above-mentioned en-
hancement effect indeed.
In conclusion we can say that our BCS-like
theory for bosons in conjunction with the IBM2
Hamiltonian suggests that open-shell nuceli (open
in protons and neutrons) may form a superfluid
condensate of a particles rather than separate
superfluid phases of proton and neutron pairs.
This is of course to be understood in the context
that finite systems never undergo a real phase
transition, but this problematic point is the same
for the condensation of a particles as it is for
the one of neutron or proton pairs; the well-
known answers to the latter case 10 apply there-
fore equally to the case of a particles.
Of course further studies on our problem are
in order. Firstly, one should perhaps take a
more general Hamiltonian (inclusion of, e.g., the
gboson is easy). Secondly, preliminary studies
1238
LETTERS 3 OCTOBER 1983
show that in the filling in of a particles the s y s -
tem has a tendency to deform; it would there-
fore be very interesting to study the phase t r a n -
sition from sphericity to deformation within our
model. Thirdly, one should try to reformulate
our theory using real fermions and not bosons
since the underlying bifermion structure of the
bosons probably plays an important role for this
problem. Investigations on these points are in
progress.
We are grateful to Ph. Nozieres for discussions
and for his constant interest in this work. This
work was supported in part by the Budesminister-
ium fur Forschung und Technologic
( a ) P e r m a n e n t a d d r e s s : P h y s i c s D e p a r t m e n t , Indian
Institute of Technology, Bombay 40076, India.
*A. A r i m a and V. Gillet, Ann. P h y s . (N.Y.) 66, 117
(1971); M. Danos and V. Gillet, Z. P h y s . 249, 294
(1972).
2
J . E i c h l e r and M. Y a m a m u r a , Nucl. P h y s . A182, 33
(1972).
3
M. K a m i m u r a , T . M a r u m o r i , and K. T a k a d a , P r o g .
T h e o r . P h y s . , Suppl. 52, 282 (1972).
4
F . Iachello and A. D. J a c k s o n , P h y s . L e t t . 108B„ 151
(1982).
5
G. G. D u s s e l , R. J . Liotta, and R. P e r a z z o , Nucl.
P h y s . A388, 606 (1982).
6
A. A r i m a and F . Iachello, Ann. Rev. N u c l . P a r t . Sci.
3 1 , 75 (1981).
7 v
' P h . N o z i e r e s and D . S a i n t - J a m e s , to be published.
8
B . H. F l o w e r s and M. Vujcic, Nucl. P h y s . 49, 586
(1963).
9
P . Schuck and S. Ethofer, Nucl. P h y s . A212, 269
(1973).
10
P. Ring and P . Schuck, The Nuclear Many Body Prob-
lem ( S p r i n g e r - V e r l a g , B e r l i n , 1980).
n
A. H. W a p s t r a and K. B o s , At. Data Nucl. Data T a -
b l e s 19, 177 (1977).