
3
62] [63]. An important point to be noted here is that
because of the positive definite nature of the term, the
sign of λdecides the sense of the active torque, making
the system naturally chiral. However, in the presence
of either polar or nematic activity, it is possible to have
linear active terms. It is also possible to model active
torques in the form of torque dipoles [64–67].
Elastostatics of odd Cosserat materials — Using the
stress tensor in Eq.(1), we now discuss the static proper-
ties of odd Cosserat materials. Since we are considering
the static regime, all time derivatives drop out. In static
equilibrium, solids balance external stresses by the elas-
tic stresses, which are proportional to the strain. In the
case of Cosserat solids, there is an additional degree of
freedom, which is the angle φ. This force balance can be
written as a set of linear equations with the applied stress
on one side and the internal stress on the other side. If
we now additionally neglect the higher order spatial gra-
dients arising due to diffusion we then have a linear prob-
lem where we can choose a profile of external stress and
from that obtain the strain in static equilibrium. Ma-
terial properties like the Young’s modulus (E), Poisson
ratio (ν), and odd ratio (νo, defining the transverse tilt
of the solid under uniaxial compression, see Ref. [68]) can
be computed using this method. We find the emergence
of auxetic properties (negative value of ν) in the limit
of large odd elasticity (2(κo)2> µB), which has been
previously reported for non-Cosserat odd elastic solids in
Ref. [68]. Details of this computation is provided in [59].
While the moduli (E,ν, and νo) remain largely un-
affected under the application of a uniaxial pressure, we
can obtain generic expressions of strain in the presence
of applied stress. Let us consider a problem where we
have only applied rotational and transverse stresses, i.e.,
only σxy and σyx are non-zero. Under such an external
stress, components of the strain tensor have the form:
∂ux
∂x =−κo(v1+v2)
4κo2+µ2,
∂uy
∂y =κo(v1+v2)
4κo2+µ2,
∂uy
∂x =v2−v1
ε+v2−v1
κc+µ(v1+v2)
2(4κo2+µ2),
∂ux
∂y =−v2−v1
ε−v2−v1
κc+µ(v1+v2)
2(4κo2+µ2),(4)
where, v1=σxy,v2=σyx, and εis a regularizing term
added to the equation of motion of φto make the matrix
invertible. Physically, one can think of εas an effective
renormalization of the higher order spatial gradients that
have been neglected and also from coupling to external
substrate.
Elastodynamics of odd Cosserat materials — We
now consider the dynamics of the system described
by Eq. (1) and Eq. (2). In the under-damped limit
(i.e., Γ →0, Γφ→0, and ignoring other viscous ef-
fects), we can obtain oscillatory solutions from the elas-
tic terms. We now consider propagating waves of the
form exp(i(k·x−ωt)), where xis the position coordi-
nate and kis the wavevector which gives us wavenumber
k=|k|. The frequency is given by ωand tis time. In the
limit of large odd elasticity, we obtain a dispersion rela-
tion: ω=eiπ/4kpκo/ρ, where, ωis the frequency and k
is the wavenumber. The above dispersion relation corre-
sponds to a propagating wave in displacement with speed
pκo/2ρand a damping constant pκo/2ρ. In the case of
a normal elastic solid without any Cosserat coupling or
odd elasticity, we have the usual elastic waves with dis-
persion relations ω=kpµ/ρ and ω=kp(B+µ)/ρ.
Therefore, unlike normal elastic waves, in odd elastic
solids, we have spontaneous damping and injection of
energy in the form of linear displacement fluctuations
within the solid.
We now consider the over-damped regime, where we
can neglect inertial effects and absorb the damping coef-
ficients (Γ and Γφ) into the other coefficients of the prob-
lem. In typical elastic materials, such an over-damped
limit gives rise to damped modes due to the effects of the
shear modulus and bulk modulus. Odd elasticity gives
rise to propagating waves, which are analogous to the
Avron waves in odd-viscous fluids [48, 69, 70]. However,
for an odd solid these are waves in displacement and not
in velocity. We now compute the dispersion relation for
a Cosserat solid in the presence of odd elasticity. For the
purpose of this calculation, we ignore the bulk modulus
and consider the limit where we can neglect the fluc-
tuations in φ, i.e., φ=φ0. We obtain a closed form
expression of the dispersion relation given by:
ω=ik2
8−8µ−κc±pκc2−64κo2,(5)
which is consistent with our understanding that the
Cosserat term is predominantly a diffusive term in the
equations of elastodynamics, whereas odd elasticity gives
rise to wave solutions with speed proportional to √κo.
Note that we obtain an exceptional point at κc= 8κo,
where the eigenvalues are degenerate and have a square
root branch point [52, 53]. At the exceptional point, two
eigenvectors coalesce to a single one and the eigenvalues
are degenerate. At this exceptional point, we find the
transition from a diffusive solution with diffusivity pro-
portional to √κcto a damped wave solution with speed
proportional to √κo. This is the key differentiating fea-
ture of the odd Cosserat elasticity from both equilibrium
elastic solids and active odd elastic solids.
For the usual Cosserat solids with κo= 0, the excep-
tional points are absent because of the Hermitian nature
of the matrix. The discriminant of the cubic equation
does not take negative values because it can be written
as a sum of positive numbers. However, in the more gen-
eral case, if we take into account the effect of both κcand
κowe obtain ω=iΛ that are solutions to the cubic equa-