
2
which in both studies did not go beyond ∼100 TeV.
While increasing the multiplicity of the fields cou-
pled to the higgs field helps to alleviate various con-
straints [12–14], in particular those related to the stabil-
ity of the scalar potential and perturbativity, this feature
is not a proximate cause of the high-temperature EW
SNR phenomenon. In this paper, we show that the addi-
tion of one scalar field coupled to the higgs is sufficient to
achieve EW SNR if the scalar develops a sufficiently large
vev at high temperatures in the early universe. Note that
increasing the scalar vev neither destabilizes the scalar
potential nor exacerbates the running of couplings. We
demonstrate this idea of realizing EW SNR via a vev in
a simple model of a complex scalar singlet coupled to
the EW sector through the higgs portal with negative
coupling. In the presence of a sufficiently large chemical
potential, the thermal equilibrium state of the new scalar
includes a Bose-Einstein condensate (BEC) [24,25] and
this condensate yields the requisite large negative higgs
mass squared for EW SNR.
Chemical potentials in the universe naturally arise in
the presence of net background charges associated with
some global symmetries. In fact, current observations
are consistent with the universe possessing large back-
ground charges of certain kinds. While the baryon asym-
metry of the universe has been observed to be tiny,
nB/s ∼10−10 [26,27], up to O(1) total lepton asym-
metry [28] is still allowed. Charge asymmetries may also
reside in the dark sector [29] at an unconstrained level.
Global symmetries are expected to be broken at high en-
ergies by higher dimensional operators [30,31]. Thus,
a field whose Lagrangian respects a global symmetry at
low energies could carry a net charge as an after effect of
its high-energy dynamics. A concrete example of this is
the Affleck-Dine mechanism [32]. Furthermore, if some
form of entropy production [33–35] or charge washout
[36–38] took place, these charge asymmetries could be
much greater in the early universe and have stronger im-
pacts then.
In this paper we show that EW SNR can be minimally
realized by coupling the higgs to a scalar that develops a
vev in the early universe. We elaborate this point further
in section II. In section III, we present a simple example
model (with a new complex scalar that forms a BEC)
that demonstrates this idea, analyze the viable parameter
space for achieving EW SNR, and describe its cosmology.
Finally, we conclude in section IV.
II. HIGH-TEMPERATURE ELECTROWEAK
SYMMETRY NON-RESTORATION WITH A
SCALAR CONDENSATE
The higgs doublet Hcan be expanded in the unitary
gauge as
H(x) = 1
√20
H+h(x).(1)
where only the real part of the neutral component has
a constant background value Hand the physical higgs
boson is denoted by h. At the minimum adopted by the
universe we have H=vH, where vHis the higgs vev. The
EW symmetry is broken in the early universe if the scalar
effective potential has no minimum in which Hvanishes.
A sufficient condition for EW SNR is the effective mass
squared m2
H(T) of the higgs field being negative at the
field space points where H= 0, i.e.
m2
H(T) = ∂2V(T, H)
∂H2H=0
<0.(2)
Here V(T, H) denotes the finite-temperature effective po-
tential evaluated using the traditional background field
method [39], which could also be a function of additional
background fields. At finite temperatures, m2
H(T) ac-
quires large positive contributions from the SM fields cou-
pled to it, leading to the usual expectation of EW symm-
metry restoration within the SM. All these SM particles
contribute positively to m2
H(T) because the fermion and
gauge-boson contributions are quadratic in their Yukawa
and gauge couplings to the higgs, respectively. For the
same reason, the EW symmetry remains to be restored
in many early universe models with extended EW sec-
tors. On the other hand, new scalar fields can couple
with negative couplings to the higgs field and yield large
negative contributions to m2
H(T).
Consider, for instance, the simplest case where a real
scalar field Sis coupled to the higgs doublet field H
through a negative higgs-portal coupling −λH S |H|2S2/2,
which also couples Sto the SM thermal bath. Through
this coupling, the thermal fluctuations of Scontribute
∼ −λHS T2to m2
H(T), which tend to push the higgs field
away from the origin (i.e. the field space points where the
higgs background field vanish H= 0). In order for one
such scalar contribution to overcome the SM contribu-
tions while keeping the tree-level scalar potential V(H, S)
bounded from below, the quartic self-coupling λSof the
Sfield would need to be non-perturbatively large [12–
14]. This led to the introduction of O(100) scalars in
Refs. [12–14] in order to realize EW SNR in the early
universe, while allowing for a perturbative treatment of
the theory and keeping the tree-level potential bounded
from below.
The preceding discussion assumes that the new scalar
fields have no appreciable chemical potential, in which
case the Bose-Einstein distribution corresponds to modes
with energy Ek.Thaving O(1) occupation numbers. In
the following we will consider more general momentum
distributions. Schematically, the contribution to m2
H(T)
from a scalar with arbitrary occupation numbers fkis
proportional to Rd3k fk/Ek. This contribution is max-
imized for a given energy density ρ∼Rd3k fkEkwhen
the occupation number fkis concentrated in the infrared
momentum modes, where the particle energy Ekis mini-
mized. Given its low entropy, it appears that a strongly-
coupled field with an IR-concentrated momentum distri-
bution would not last for a long time. However, such