Derrick's theorem

Original argument

Derrick's paper, which was considered an obstacle to interpreting soliton-like solutions as particles, contained the following physical argument about non-existence of stable localized stationary solutions to the nonlinear wave equation

now known under the name of Derrick's Theorem. (Above, f ( s ) is a differentiable function with f ′ ( 0 ) = 0 .)

The energy of the time-independent solution θ ( x ) is given by

A necessary condition for the solution to be stable is δ 2 E ≥ 0 . Suppose θ ( x ) is a localized solution of δ E = 0 . Define θ λ ( x ) = θ ( λ x ) where λ is an arbitrary constant, and write I 1 = ∫ ( ∇ θ ) 2 d 3 x , I 2 = ∫ f ( θ ) d 3 x . Then

Whence ( d E λ / d λ ) | λ = 1 = − I 1 − 3 I 2 = 0 , and since I 1 > 0 ,

That is, δ 2 E < 0 for a variation corresponding to a uniform stretching of the particle. Hence the solution θ ( x ) is unstable.

The above argument also works for x ∈ R n , n > 3 .

Pohozaev's identity

More generally, let g be continuous, with g ( 0 ) = 0 . Denote G ( t ) = ∫ 0 t g ( s ) d s . Let

be a solution to the equation

in the sense of distributions. Then u satisfies the relation

known as Pohozaev's identity. It result is similar to the Virial theorem.

Interpretation in the Hamiltonian form

We may write the equation ∂ t 2 u = ∇ 2 u − 1 2 f ′ ( u ) in the Hamiltonian form ∂ t u = δ v H ( u , v ) , ∂ t v = − δ u H ( u , v ) , where u , v are functions of x ∈ R n , t ∈ R , the Hamilton function is given by

and δ u H , δ v H are the variational derivatives of H ( u , v ) .

Then the stationary solution u ( x , t ) = θ ( x ) has the energy H ( θ , 0 ) = ∫ R n ( 1 2 | ∇ θ | 2 + 1 2 f ( θ ) ) d n x and satisfies the equation

with E ′ denoting a variational derivative of the functional E = ∫ R n [ | ∇ θ | 2 + f ( θ ) ] d n x . Although the solution θ ( x ) is a critical point of E (since E ′ ( θ ) = 0 ), Derrick's argument shows that d 2 d λ 2 E ( θ ( λ x ) ) < 0 at λ = 1 , hence u ( x , t ) = θ ( x ) is not a point of the local minimum of the energy functional H . Therefore, physically, the solution θ ( x ) is expected to be unstable.

Stability of localized time-periodic solutions

Derrick describes some possible ways out of this difficulty, including the conjecture that Elementary particles might correspond to stable, localized solutions which are periodic in time, rather than time-independent. Indeed, it was later shown that a time-periodic solitary wave u ( x , t ) = ϕ ω ( x ) e − i ω t with frequency ω may be orbitally stable if the Vakhitov–Kolokolov stability criterion is satisfied.