taining large laser-driven accelerations despite the inevitable
presence of perturbations (see Fig. 1). These perturbations—
arising from multiple possible sources such as atmospheric dis-
turbances, beam non-uniformities, shape distortion of the sail,
etc.—prevent perfectly uniform loading of the sail, and this
complication of the laser-lightsail dynamics may cause the dis-
tortions in sail shape to grow in amplitude. Even if the per-
turbations are, through careful lightsail deployment, kept small
in magnitude, the following question remains: Under the large
photonic pressure loads required for feasible interstellar flight,
will the structural perturbations of the lightsail grow out of
bound or will the lightsail remain flat?
The present study addresses the question of lightsail shape
stability with perturbations by considering a first-principles ap-
proach to the problem. First, a continuous model of a sinu-
soidally perturbed lightsail under radiation load allows an ana-
lytical, quasi-static analysis of the critical point between light-
sail shape stability and instability in Section 2. A Lagrangian-
based finite element (FE) numerical model of the perturbed
lightsail is then constructed using rigid rectangular slices to
simulate the full dynamics of the lightsail while undergoing ac-
celeration in Section 3. The rigid-element numerical model was
further generalized to include torsion and rectilinear springs
to investigate the influence upon lightsail structural stability of
material bending stiffness and applied tension, respectively. In
Section 4, the quasi-static derived analytical expressions are
compared to the numerical results, which consisted of multi-
ple lightsail dynamics simulations with each simulation vary-
ing lightsail geometry, material modulus, and/or applied ten-
sion magnitude. The engineering implications of the analysis
are then explored.
2. Theoretical Considerations
As a preliminary consideration of the problem of a lightsail
under radiation loading, an L×W×hcontinuous elastic plate
model was first constructed that would allow for a static anal-
ysis of the criticality between a structurally stable and unstable
lightsail (see Fig. 1). The 3-dimensional plate model was then
simplified to a 2-dimensional beam model by setting the light-
sail width, W, equal to unit length (in meters). The continuous
elastic beam model is shown in Fig. 2. The following analysis
considers introducing a sinusoidal perturbation into the lightsail
shape. The restoring bending moment caused by this deforma-
tion is then compared to the moment induced by the incident ra-
diation interacting with the curve surface of the lightsail. If the
radiation-induced moment exceeds the bending moment asso-
ciated with the imposed perturbation and acts in the same direc-
tion as the perturbation, then presumably the lightsail continues
to further deform. If the radiation-induced moment is less than
the restoring bending moment of the perturbation, then the sail
would be expected to return toward its original, flat configura-
tion. By equating the bending moment of the sinusoidal pertur-
bation with the radiation-induced moment, a critical condition
for lightsail stability can be defined in terms of the radiation
intensity, the elastic modulus and dimensions of the lightsail,
and the amplitude and wave number of the perturbation. The
formalism of this approach follows here.
The beam model analysis is here conducted in a non-inertial
reference frame accelerating at g0, the lightsail’s vertical ac-
celeration, with a body force term ρh W g0acting in the y-
direction on the mass elements of the lightsail in accordance
with D’Alembert’s principle, thereby reducing the dynamic prob-
lem to a quasi-static problem.3The presence of radiation pres-
sure is modeled by the distributed loading, pr. The expression
of prcan be derived by first considering the kinematics and
dynamics of an infinitesimal flat sail element of length dsas
shown in Fig. 3 for reference. For a lightsail element made of
a material with reflection coefficient r, absorption coefficient a,
and transmission coefficient ˜τ, the total force imparted by ra-
diation of oblique incidence angle θcan be resolved into the
normal-tangential components
fn=I0
c"(1 +rs) cos2θ+Bf(1 −s)rcos θ
+(1 −r−˜τ)εfBf+εbBb
εf+εb
cos θ#n,
(1)
ft=I0
c(1−rs)cos θsin θt,(2)
where sstands for the fraction of light that is specularly re-
flected; εfand εbstand for the emissivity of the front (reflect-
ing) and back (non-reflecting) surfaces of the sail, respectively;
and Bfand Bbare the coefficients accounting for the poten-
tially non-Lambertian nature of the sail element surfaces. To-
gether, these two forces generate a resultant whose direction,
m, is skewed away from the element’s normal:
f=qf2
n+f2
tm.(3)
This optical lightsail model was first proposed by Forward [45]
and then was further discussed by Wright and McInnes [46,
47] now termed the Forward-Wright-McInnes or FWM model.4
For the purposes of this paper, the lightsail element will be as-
sumed to be perfectly reflective (r=1=⇒a=˜τ=0) with all
beam reflections being specular (s=1). The resulting force per
element then becomes
f=2I0
cAcos2θn,(4)
a force that is entirely normal to the sail element. Normaliz-
ing (4) with respect to the element area generates the radiation
pressure:
pr=2I0
ccos θ. (5)
3Given the large magnitude of the vertical accelerations, it is here assumed
that g0=¨y¨x,¨
θ(s), and thus the horizontal and rotational inertia of the light-
sail were deemed negligible and their d’Alembert equivalent was not included
into the lightsail beam model.
4The authors are aware that more accurate lightsail optical models using
vector theories such as the Rayleigh-Rice theory have been proposed [48, 49],
but the FWM model was used here to ensure a straightforward simplification to
the case of an ideal reflector. The implementation of physically more accurate
optical models can be incorporated into future studies.
4