Abstract
We study the spectral stability of the nonlinear Dirac operator in dimension \(1+1\), restricting our attention to nonlinearities of the form \(f(\left\langle \psi ,\beta \psi \right\rangle _{\mathbb {C}^2}) \beta \). We obtain bounds on eigenvalues for the linearized operator around standing wave solutions of the form \(e^{-i\omega t} \phi _0\). For the case of power nonlinearities \(f(s)= s |s|^{p-1}\), \(p>0\), we obtain a range of frequencies \(\omega \) such that the linearized operator has no unstable eigenvalues on the axes of the complex plane. As a crucial part of the proofs, we obtain a detailed description of the spectra of the self-adjoint blocks in the linearized operator. In particular, we show that the condition \(\left\langle \phi _0,\beta \phi _0 \right\rangle _{\mathbb {C}^2} > 0\) characterizes groundstates analogously to the Schrödinger case.
Similar content being viewed by others
Data Availability
Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.
Notes
According to numerical work in [14], this type of eigenvalue branches do occur in higher dimensions for non-radial perturbations.
Analytic functions of \(\mu \) as long as eigenvalues are simple, and branches of a multivalued function analytic in \((\mu -\mu _0)^{1/k}\) if at least \(k \in \mathbb {N}\) branches intersect for \(\mu _0\).
Up to the additional term in the denominator.
Also called Ferrers functions, since \(-1<\tanh <1\).
References
Berezin, F.A., Shubin, M.A.: The Schrödinger Equation, Mathematics and Its Applications, Soviet Series, vol. 66, 1st edn. Springer, Dordrecht (1991)
Berkolaiko, G., Comech, A.: On spectral stability of solitary waves of nonlinear Dirac equation in 1D. Math. Model. Nat. Phenom. 7, 13–31 (2012)
Berkolaiko, G., Comech, A., Sukhtayev, A.: Vakhitov–Kolokolov and energy vanishing conditions for linear instability of solitary waves in models of classical self-interacting spinor fields. Nonlinearity 28, 577–592 (2015)
Boussaïd, N., Cacciapuoti, C., Carlone, R., Comech, A., Noja, D., Posilicano, A.: Spectral stability and instability of solitary waves of the Dirac equation with concentrated nonlinearity, arXiv:2006.03345 (2020)
Boussaïd, N., Comech, A.: On spectral stability of the nonlinear Dirac equation. J. Funct. Anal. 271, 1462–1524 (2016)
Boussaïd, N., Comech, A.: Nonlinear Dirac Equation. Mathematical Surveys and Monographs, vol. 244. American Mathematical Society, Providence (2019)
Boussaïd, N., Comech, A.: Spectral stability of small amplitude solitary waves of the Dirac equation with the Soler-type nonlinearity. J. Funct. Anal. 277, 108289, 68 (2019)
Boussaïd, N., Cuccagna, S.: On stability of standing waves of nonlinear Dirac equations. Commun. Partial Differ. Equ. 37, 1001–1056 (2012)
Cazenave, T., Vázquez, L.: Existence of localized solutions for a classical nonlinear Dirac field. Commun. Math. Phys. 105, 35–47 (1986)
Chugunova, M., Pelinovsky, D.E.: Block-diagonalization of the symmetric first-order coupled-mode system. SIAM J. Appl. Dyn. Syst. 5, 66–83 (2006)
Comech, A., Guan, M., Gustafson, S.: On linear instability of solitary waves for the nonlinear Dirac equation. Ann. Inst. H. Poincaré Anal. Non Linéaire 31, 639–654 (2014)
Cooper, F., Khare, A., Mihaila, B., Saxena, A.: Solitary waves in the nonlinear Dirac equation with arbitrary nonlinearity. Phys. Rev. E (3) 82, 036604, 14 (2010)
Cuccagna, S.: The Hamiltonian structure of the nonlinear Schrödinger equation and the asymptotic stability of its ground states. Commun. Math. Phys. 305, 279–331 (2011)
Cuevas-Maraver, J., Boussaïd, N., Comech, A., Lan, R., Kevrekidis, P.G., Saxena, A.: Solitary Waves in the Nonlinear Dirac Equation, pp. 89–143. Springer (2018)
Dirac, P.A.M.: The quantum theory of the electron. Proc. R. Soc. Lond. A 117, 610–624 (1928)
Dolbeault, J., Esteban, M.J., Séré, É.: On the eigenvalues of operators with gaps. Application to Dirac operators. J. Funct. Anal. 174, 208–226 (2000)
Dolbeault, J., Esteban, M.J., Séré, É.: General results on the eigenvalues of operators with gaps, arising from both ends of the gaps. Application to Dirac operators. J. Eur. Math. Soc. (JEMS) 8, 243–251 (2006)
Esteban, M.J., Lewin, M., Séré, É.: Domains for Dirac–Coulomb min–max levels. Rev. Mat. Iberoam. 35, 877–924 (2019)
Griesemer, M., Siedentop, H.: A minimax principle for the eigenvalues in spectral gaps. J. Lond. Math. Soc. (2) 60, 490–500 (1999)
Grillakis, M., Shatah, J., Strauss, W.: Stability theory of solitary waves in the presence of symmetry. I. J. Funct. Anal. 74, 160–197 (1987)
Gross, D.J., Neveu, A.: Dynamical symmetry breaking in asymptotically free field theories. Phys. Rev. D 10, 3235 (1974)
Hislop, P., Sigal, I.M.: Introduction to Spectral Theory with Application to Schrödinger Operators. Springer, New York (1996)
Ivanenko, D.D.: Notes to the theory of interaction via particles. Zh. Éksp. Teor. Fiz 8, 260–266 (1938)
Kato, T.: Perturbation Theory for Linear Operators, 2nd edn. Springer (1995)
Kolokolov, A.A.: Stability of the dominant mode of the nonlinear wave equation in a cubic medium. J. Appl. Mech. Tech. Phys. 14, 426–428 (1973)
Lakoba, T.I.: Numerical study of solitary wave stability in cubic nonlinear Dirac equations in 1D. Phys. Lett. A 382, 300–308 (2018)
Lee, S.Y., Kuo, T.-K., Gavrielides, A.T.: Exact localized solutions of two-dimensional field theories of massive fermions with Fermi interactions. Phys. Rev. D 12, 2249–2253 (1975)
Mertens, F.G., Quintero, N.R., Cooper, F., Khare, A., Saxena, A.: Nonlinear Dirac equation solitary waves in external fields. Phys. Rev. E 86, 046602 (2012)
Pauli, W.: Contributions mathématiques à la théorie des matrices de Dirac. Ann. Inst. Henri Poincaré 6, 109–136 (1936)
Pelinovsky, D.E., Shimabukuro, Y.: Orbital stability of Dirac solitons. Lett. Math. Phys. 104, 21–41 (2014)
Pelinovsky, D.E., Stefanov, A.G.: Asymptotic stability of small gap solitons in nonlinear Dirac equations. J. Math. Phys. 53, 073705 (2012)
Reed, M., Simon, B.: Methods of Modern Mathematical Physics. IV. Analysis of Operators. Academic Press [Harcourt Brace Jovanovich Publishers], New York, London (1978)
Schimmer, L., Solovej, J.P., Tokus, S.: Friedrichs extension and min–max principle for operators with a gap. Ann. Henri Poincaré 21, 327–357 (2020)
Soler, M.: Classical, stable, nonlinear spinor field with positive rest energy. Phys. Rev. D 1, 2766–2769 (1970)
Thaller, B.: The Dirac Equation, Texts and Monographs in Physics. Springer, Berlin (1992)
Vakhitov, N.G., Kolokolov, A.A.: Stationary solutions of the wave equation in a medium with nonlinearity saturation. Radiophys. Quantum Electron. 16, 783–789 (1973)
Weinstein, M.I.: Modulational stability of ground states of nonlinear Schrödinger equations. SIAM J. Math. Anal. 16, 472–491 (1985)
Acknowledgements
We thank Sébastien Breteaux, Jérémy Mougel, and Phan Thành Nam for helpful discussions, and Matías Moreno for checking some of the computations. We thank Andrew Comech for his remarks on the preprint version. E.S. thanks Thomas Sørensen and the Center for Advanced Studies at LMU for their hospitality during his stay in Munich where part of this work took place. We thank the referees for many suggestions that improved the manuscript.
Funding
The research visits leading to this work where partially funded by ANID (Chile) Project REDI–170157. J.R. received funding from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy (EXC-2111-390814868). E.S. and H.VDB. have been partially funded by ANID (Chile) through Fondecyt Project #118-0355. H.VDB. acknowledges support from ANID through Fondecyt Projects #318-0059 and #11220194 and from CMM through ANID PIA AFB17000, #ACE210010 and Project France-Chile MathAmSud EEQUADDII 20-MATH-04.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Communicated by K. Nakanishi.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix A. ODE arguments
In this appendix, we give the proofs of Propositions 2.1 (basic properties of the groundstates) and 2.5 because some intermediate steps are useful in both proofs. However, we stress that the proof of Proposition 2.5 requires spectral properties of \(L_2\) that have been established in Proposition 2.2 and Lemma 2.3.
Proof of Proposition 2.1
Equation (3) satisfied by \(\phi _0(\omega ) = (v_\omega , u_\omega )^{\textsf{T}}\) can be written as
with \(M_\omega = m - f\!\left( v_\omega ^2 - u_\omega ^2\right) \) as defined in (9). Therefore, we assume \(v_\omega (0)\geqslant 0\) since \(-\phi _0(\omega )\) is a solution if and only if \(\phi _0(\omega )\) is a solution. We define \(F(s) := \int _0^s f(t) \,\textrm{d}t \) on \((0,+\infty )\) and \({\tilde{F}}\in {\mathcal {C}}^0([0,+\infty ))\cap {\mathcal {C}}^1((0,+\infty ))\) by \({\tilde{F}}(0)=0\) and \({\tilde{F}}(s) := F(s)/s\) on \((0,+\infty )\). The continuity at the origin derives from \(f'>0\) on \((0,+\infty )\) by Assumption 1.1, as it gives
where we used the assumption \(f(0)=0\) for the positivity of F.
Existence, positivity and regularity. The existence is established in [2, Lemma 3.2]. In its notation, we have \(g(s) = m - f(s)\) and \(G(s) = m s - F(s)\), and have to check that
First, \(G(s)/s>g(s)\) on \((0,+\infty )\) by (51)—hence checking already, for any \(s^*_\omega >0\), the non-equality—, which implies \(\lim _{s\searrow 0} G(s)/s \geqslant g(0) = m > \omega \). Second, denoting \(f^{-1}\) the inverse of \(f:[0,+\infty )\rightarrow [0,\lim _{+\infty } f)\), \(\lim _\infty f \geqslant m\) from Assumption 1.1 gives
Thus \(\lim _{+\infty } {\tilde{F}} \geqslant m\), and \(\lim _{s\rightarrow +\infty } G(s)/s \leqslant 0 < \omega \). Third, \(s^2 {\tilde{F}}'(s)= s f(s) - F(s)>0\) on \((0,+\infty )\) by (51). Summarizing, \(G(s)/s = m - {\tilde{F}}(s)\) is strictly decreasing on \((0,+\infty )\) with \(\omega \) is in its image set, hence there is a unique \(s^*_\omega >0\) satisfying the properties. The existence of a non-zero solution \((v_\omega , u_\omega ) \in H^1(\mathbb {R},\mathbb {R}^2)\) to (3) with \(v_\omega \) is even and \(u_\omega \) odd is therefore proved by [2, Lemma 3.2]. Moreover, since \(u_\omega (0)=0\) and \(v_\omega (0)\geqslant 0\), we can assume that \(v_\omega (0)>0\) as, otherwise, \(\phi _0(\omega )\) is the trivial solution by the Cauchy theory.
To prove (i), notice that (50) is the Hamiltonian system \(h_\omega (u_\omega ,v_\omega )=0\) associated to
On one hand, since \(f>{\tilde{F}}\) on \((0,+\infty )\) by (51) and \(v_\omega (0)>0\), (50) implies
On another hand, it also gives \(u_\omega (x) \ne 0\) if \(x \ne 0\) since, otherwise, there would be \(x >0\) (by oddity of u) such that \((v_\omega (x), u_\omega (x))= (v_\omega (0), u_\omega (0))\) and the solution would be periodic (in x), contradicting \(\phi _0(\omega ) \in L^2(\mathbb {R})\). Hence, we conclude that \(u_\omega >0\) on \((0,+\infty )\), since \(u_\omega \) is continuous, non-zero on \((0,+\infty )\) with \(u_\omega '(0)>0\) and \(u_\omega (0)=0\).
Moreover, \(d_\omega :=v_\omega ^2 - u_\omega ^2>0\) on \(\mathbb {R}\). Indeed, \(d_\omega (0)=v_\omega (0)^2>0\) and, if \(d_\omega (x)=0\) for some \(x\ne 0\), then \(0=h_\omega (u_\omega ,v_\omega )(x) = \frac{\omega }{2} (v_\omega (x)^2 + u_\omega (x)^2)\) hence \(v_\omega (x)= u_\omega (x) = 0\) contradicting \(u_\omega \ne 0\) on \(\mathbb {R}{\setminus }\{0\}\) just obtained. Finally, \(v_\omega> u_\omega >0\) on \((0,+\infty )\) because \(|v_\omega |> u_\omega >0\) on \((0,+\infty )\), since \(d_\omega >0\) on \(\mathbb {R}\), with \(v_\omega \) continuous and \(v_\omega (0)>0\).
Since \((v_\omega , u_\omega )\in H^1(\mathbb {R})\subset {\mathcal {C}}^0(\mathbb {R})\) and \(f\in {\mathcal {C}}^0(\mathbb {R})\) by Assumption 1.1, (50) gives \((v_\omega , u_\omega )\in {\mathcal {C}}^1(\mathbb {R})\). Moreover, \(f\circ d_\omega \in {\mathcal {C}}^1(\mathbb {R})\) since \(d_\omega >0\) on \(\mathbb {R}\) and \(f\in {\mathcal {C}}^1(\mathbb {R}{\setminus }\{0\})\) by Assumption 1.1, hence (50) actually gives \((v_\omega , u_\omega )\in {\mathcal {C}}^2(\mathbb {R})\).
Uniqueness and continuity in \(\omega \) of the initial condition. We have
Moreover, we already prove that \({\tilde{F}}\in {\mathcal {C}}^0([0,+\infty ))\cap {\mathcal {C}}^1((0,+\infty ))\) is strictly increasing on \([0,+\infty )\) with \(\lim _{+\infty } {\tilde{F}} \geqslant m\). Thus, it is bijective from \([0,+\infty )\) to \([0,\lim _{+\infty } {\tilde{F}})\) and has a continuous inverse with \(v_\omega (0)^2 = {\tilde{F}}^{-1}(m-\omega )>0\). By the Cauchy theory, this implies the uniqueness of \(\phi _0(\omega )\).
Pointwise decay of \(\phi _0(\omega ,\cdot )\) as \(\omega \rightarrow m\). We start by noticing that \(h_\omega (u_\omega ,v_\omega ) = 0\) gives
due to \(F >0\) on \((0,+\infty )\), \(v_\omega ^2(0)> 0 = u_\omega ^2(0)\), and the parities of \(v_\omega \) and \(u_\omega \). Moreover, \(\left| \left| d_\omega \right| \right| _{L^\infty } = d_\omega (0)\), because \((d_\omega )' = - 4 \omega u_\omega v_\omega \) by (50) and \(v_\omega>u_\omega >0\) on \((0,+\infty )\), hence
for \(\omega \in (0,m)\). Since \({\tilde{F}}^{-1}(0) = 0\), this establishes half of (iv).
Uniform exponential decay of \(\phi _0(\omega ,\cdot )\). In order to prove differentiability, it is important to obtain a pointwise upper bound uniform in \(\omega \), for \(\omega \) bounded away from m and 0. Thus, for fixed \(\epsilon > 0\), we assume that \(m^2-\omega ^2 \geqslant \epsilon ^2 > 0\). In view of (52), it is sufficient to bound \(d_\omega \), and by symmetry we assume that \(x \geqslant 0\). We have
using \(h(v_\omega , u_\omega ) = 0\) for the last equality. Since \(\lim _{s\searrow 0} (m - {\tilde{F}}(s)) \geqslant m > \omega \) (see the proof of existence and remember that \(m - {\tilde{F}}(s)=G(s)/s\)), we define \(s_\epsilon \) such that
Since \(d_\omega \) is strictly decreasing on \([0,+\infty )\) hence bijective on it, we define \(x^*(\omega ,\epsilon )\) by \(x^*(\omega ,\epsilon ) = 0\) if \(d_\omega (0) \leqslant s_\epsilon \) and \(x^*(\omega ,\epsilon ) = d_\omega ^{-1}(s_\epsilon )\) otherwise, and we have
Integrating it, yields (ii) since it gives
Decays of Q. By the definition of Q, the positivity of \(d_\omega \), the one of \(f'\) on \((0,+\infty )\) from Assumption 1.1, and (52), we bound
It yields (iii), the exponential decay in x, due to the exponential decay of \(v_\omega \) and \(u_\omega \), and to \(\lim _{s\rightarrow 0^+} s f'(s) = 0\) from Assumption 1.1. It also gives the statement on Q in (iv):
\(\square \)
Proof of Proposition 2.5
We freely use the notation introduced in the previous proof. We remind the reader that the following proof of differentiability of \(\omega \mapsto \phi _0(\omega )\) is needed because we do not assume continuity of \(f'\) at 0.
Uniform continuity in \(\omega \). We show that \(\omega \mapsto \phi _0(\omega )\) is continuous with values in \({\mathcal {C}}^0(\mathbb {R})\). We rewrite the nonlinear equation (3) as an initial value problem on \([0, +\infty )\) of the form
with
Here, \(B(\omega , \phi )\) is of class \({\mathcal {C}}^1\) in \(\omega \) and \(\phi \) (as long as \(\phi _1^2 - \phi _2^2\) is bounded away from zero, which is the case on any bounded interval), and thus \(\omega \mapsto \phi _0(\omega ,x)\) is continuous in \(\omega \), uniformly for x in bounded intervals.
We now use the exponential decay from the previous proof. The pointwise continuity implies that \(x^*(\omega , \epsilon )\) is bounded for \(\omega \in [\epsilon , \sqrt{m^2-\epsilon ^2}]\), since otherwise there would be \(\omega ^*\) such that \(d_{\omega ^*}(x) > s_\epsilon \) for all \(x \geqslant 0\). Thus, there exist \(r_\epsilon \) and \(A_\epsilon \) such that
The restriction that \(\omega \geqslant \epsilon \) comes from (52). Combined with the uniform continuity on bounded intervals, this shows that \(\omega \mapsto \phi _0(\omega )\) is continuous with values in \({\mathcal {C}}^0(\mathbb {R})\).
Differentiability. We fix \(\epsilon > 0\) and \(\omega \in [\epsilon , \sqrt{m^2-\epsilon ^2}]\). Assuming \(\alpha \in [\epsilon , \sqrt{m^2-\epsilon ^2}]\), substracting the groundstate equations for \(\omega \) and \(\alpha \) gives
For fixed \(x \in \mathbb {R}\), we apply the mean value theorem to f in order to find \({{\tilde{d}}}(x)\), between \(d_\alpha (x)\) and \(d_\omega (x)\), such that
where we have defined
Inserting this in the previous identity, we obtain
We claim that \({\tilde{L}}_{2, \alpha }\) converges to \(L_2(\omega )\), in \({\mathcal {B}}(H^1(\mathbb {R}), L^2(\mathbb {R}))\), as \(\alpha \) tends to \(\omega \). For the first two terms, this is clear. For the third term, we use
and the uniform continuity from the previous proof. For the last term, we use the uniform convergence on bounded intervals (where \(d_\omega \) and \(d_\alpha \) are bounded away from zero). In order to treat large values of x, we factorize out \({\tilde{d}}\) in the definition of \({\widetilde{Q}}_\alpha \):
where the factors in parenthesis are bounded in view of (52) and their pre-factor \({\tilde{d}} f'({\tilde{d}})\) vanishes at infinity since \(\lim _{s\rightarrow 0^+} s f'(s) = 0\) by Assumption 1.1.
The convergence in \({\mathcal {B}}(H^1(\mathbb {R}), L^2(\mathbb {R}))\) implies norm resolvent convergence and, in particular, convergence of the spectrum. Recall that for each fixed \(\omega \), \(L_2(\omega )\) has zero as an isolated simple eigenvalue with an odd eigenfunction. Therefore, for sufficiently small \(\left| \alpha -\omega \right| \),
is invertible and we conclude that
This shows the differentiability of \(\omega \mapsto \phi _0(\omega )\) as a function with values in \(H^{1}(\mathbb {R},\mathbb {C}^2)\), with \(\partial _\omega \phi _0(\omega ) = \left( L_2( \omega )\right) ^{-1} \phi _0(\omega )\). This function is continuous in \(\omega \) because \(\left( L_2( \omega )\right) ^{-1}\) is bounded on \(L^{2,\textrm{even}}(\mathbb {R})\). Taking the inner product with \(\phi _0(\omega )\) concludes the proof. \(\square \)
Appendix B. Proof of Lemma 4.4
We start from (18), where we introduce the parameter \(\theta :=\frac{\mu \left| \left| \left| Q \right| \right| \right| }{4(t+\omega )}\) for shortness, and study the function
Since we do not know the value of \(\eta >0\), we need to study the question: Given \(\xi \geqslant 0\), for which \(\theta >0\) the inequality
is verified? Notice that
is a decreasing function of \(\alpha >0\) for any \(\eta >0\). Thus,
Defining \(\eta _\star \) as the positive number such that \(-\eta _\star ^2 +\theta \eta _\star +1=0\), i.e.,
we have for all \(\eta \in (0,\eta _\star ]\) that
Thus, for \(\eta \in (0,\eta _\star ]\), \(\alpha \mapsto g_\theta (\alpha ,\eta )\) is an increasing function on [0, 1] and its maximum is \(g_\theta (1,\eta ) = 2(1 - \eta \theta )\).
Similarly, for any \(\alpha >0\),
Defining \(\eta _\circ \) as the positive number such that \(-\eta _\circ ^2 +4\theta \eta _\circ +1=0\), i.e., \(\eta _\circ :=2\theta + \sqrt{4\theta ^2+1}\), we have for all \(\eta \geqslant \eta _\circ \) that
Thus, for \(\eta \geqslant \eta _\circ \), \(\alpha \mapsto g_\theta (\alpha ,\eta )\) is a decreasing function on [0, 1] and its maximum is \(g_\theta (0,\eta ) = \eta ^2 - 4\theta \eta + 1\).
Finally, for \(\eta \in (\eta _\star ,\eta _\circ )\), \(\alpha \mapsto g_\theta (\alpha ,\eta )\) is increasing then decreasing on [0, 1], and reaches its maximum at \(\alpha _+:=\sqrt{\frac{4\eta \theta }{\eta ^2-1}} - 1\) with value
Now, since \(\eta \mapsto g_\theta (0,\eta ) = \eta ^2 - 4\theta \eta + 1\geqslant 2\) on \((\eta _\circ ,+\infty )\), \(\eta \mapsto g_\theta (1,\eta ) = 2(1 - \eta \theta )\) is decreasing (\(\theta >0\)), \(h(\eta _\star ) = 2(1 - \eta _\star \theta )\) by construction, and (using the identity \(\eta _\star ^2= \theta \eta _\star + 1\))
we conclude that
where \(\eta _\theta \) is defined as the unique real number in \((1,+\infty )\) such that
Note that the l.h.s. of (53) being a strictly increasing function on \((1,+\infty )\)—hence one-to-one from \((1,+\infty )\) to \((0,+\infty )\ni \theta \)—, it gives that (53) has a unique solution \(\eta _\theta \) in \((1,+\infty )\) and, on another hand that \(\eta _\star<\eta _\theta <\eta _\circ \). Indeed, recalling that \(1<\eta _\star <\eta _\circ \) and the equations they respectively solve, we have
and
Moreover, as a by-product, the problem has a solution only for \(\xi <2\), since
Summarizing, we have obtained \(\eta _\theta >1\) defined by (53) for which
Now, keeping in mind that \(\xi \in [0,2)\), that \(\eta \mapsto 2\eta ^2\frac{3-\eta ^2}{3\eta ^2 -1}\) is strictly decreasing and that \(\eta \mapsto 4\frac{\eta _\theta ^3(\eta _\theta ^2-1)}{(3\eta _\theta ^2-1)^2}\) is stritly increasing, we have
where \(\theta _+\) is defined in (23), that is, \(\theta _+:[0,2)\rightarrow (0,3\sqrt{3}/8]\) with
\(\square \)
Appendix C. Details on Some Computations
1.1 C.1. \(L^\infty \)-norms of terms involving \({\tilde{u}}\) and \({\tilde{v}}\).
We give here the details on computing several \(L^\infty \)-norms that we need in Sect. 7.
Expansion of \(\left| \left| ({\tilde{v}}^2-{\tilde{u}}^2)^p \right| \right| _\infty \). We have
Expansion of \(\left| \left| ({\tilde{v}}^2-{\tilde{u}}^2)^{p-1}{\tilde{u}}{\tilde{v}} \right| \right| _\infty \). Defining \(h_3(y):=\frac{y\left( 1-y^2\right) }{\left( 1-\nu y^2\right) ^2}\), we have
Since \(h_3\), defined on \((-1,1)\), attains at \(\pm \sqrt{y_3}\), with \(y_3:=\frac{ - 3(1-\nu ) + \sqrt{9(1-\nu )^2+4\nu }}{2\nu }\), its extrema \(\pm \frac{\sqrt{y_3}(1-y_3)}{(1-\nu y_3)^2}\), we have the non-relativistic expansion
Expansion of \(\left| \left| ({\tilde{v}}^2-{\tilde{u}}^2)^{p-1}{\tilde{u}}^2 \right| \right| _\infty \). Since on [0, 1), \(h_1(y):=\frac{1-y}{1-\nu y}\frac{\nu y}{1-\nu y}\) is nonnegative with a maximum \(\frac{\nu }{4(1-\nu )} = \frac{m-\omega }{8\omega }\) at \(\frac{1}{2-\nu }\), we have
Expansion of \(\left| \left| ({\tilde{v}}^2-{\tilde{u}}^2)^p - \frac{(p+1)\kappa ^2}{2m\cosh ^2} \right| \right| _\infty \). Since \(h_2(y):= \frac{1}{m-\omega +2\omega y} - \frac{1}{2m y}\) is positive on \((0,+\infty )\) with a maximum \(\frac{1}{m}\frac{\sqrt{m}-\sqrt{\omega }}{\sqrt{m}+\sqrt{\omega }}\) at \(\frac{1}{2}\left( 1+\sqrt{\frac{m}{\omega }}\right) \), we have
1.2 C.2. Bounds on derivatives of \(h \in F_k\)
For fixed \({\mathfrak {s}}\) defined in (34), we denote by
the eigenfunctions of the Schrödinger operator
They satisfy the eigenvalue equation
We want to bound the first and second derivatives of \(h \in {{\,\textrm{span}\,}}\{p_1, \ldots , p_k\}\). First,
For the first bound, we multiply the eigenvalue equation by \(p_j\), integrate by parts in the first term and bound \(\cosh ^{-2}\leqslant 1\) in order to obtain
For the second bound, we take the norm on both sides of the eigenvalue equation and, since \(0<{\mathfrak {s}}+1-j\leqslant {\mathfrak {s}}\) and \(0<({\mathfrak {s}}+1-j)^2\leqslant {\mathfrak {s}}^2<{\mathfrak {s}}({\mathfrak {s}}+1)\) for \(j\in \llbracket 1, \lceil {\mathfrak {s}} \rceil \rrbracket \), we obtain for all \(j\in \llbracket 1, \lceil {\mathfrak {s}} \rceil \rrbracket \) that
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Aldunate, D., Ricaud, J., Stockmeyer, E. et al. Results on the Spectral Stability of Standing Wave Solutions of the Soler Model in 1-D. Commun. Math. Phys. 401, 227–273 (2023). https://doi.org/10.1007/s00220-023-04646-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00220-023-04646-4