Abstract
In this paper, we analyze the solubility of a class of abstract fourth-order in time linear evolution equations, using the roots of the characteristic polynomial that is associated with the equations.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Let X be a separable Hilbert space and \(A:D(A)\subset X\rightarrow X\) be an unbounded linear, closed, densely defined, self-adjoint and for some \(a>0\), satisfies \(\langle Au,u\rangle \geqslant a \langle u,u\rangle \) or all \(u\in D(A)\). We also assume that A has compact resolvent on X. It is well-known that with those hypotheses, we can define the fractional power \(A^{\alpha }\) of order \(0<\alpha <1\) according to [2] and [11], as a closed linear operator, see e.g. [10, 11, 13, 18]. Denote by \(X^{\alpha }=D(A^{\alpha })\) for \(0\leqslant \alpha \leqslant 1\) (taking \(A^0:=I\) on \(X^0:=X\) when \(\alpha =0\)). Recall that \(X^{\alpha }\) is dense in X for all \(0\leqslant \alpha \leqslant 1\), for details see [2, Theorem 4.6.5]. The fractional power space \(X^\alpha \) endowed with the norm \(\Vert \cdot \Vert _{X^\alpha }:=\Vert A^{\alpha } \cdot \Vert _X\) is a Banach space. With this notation, we have \(X^{-\alpha }=(X^\alpha )'\) for all \(\alpha >0\), see [2, 26, 27] for the characterization of the negative scale.
In this paper we consider the fourth-order linear equation in time
where \(a,b,c\geqslant 0\).
We discuss here the well-posedness of Eq. (1.1) in terms of the scalar parameters a, b and c considering the theory of strongly continuous bounded linear operators on suitable phase spaces.
In recent years higher-order evolution equations in time have attracted the attention of many researchers. This can be explained by the sensitivity of the conditions of well-posedness and regularity of solutions of these equations in infinite-dimensional Banach spaces, see e.g. [3,4,5, 8, 9, 22], see also the Moore-Gibson-Thompson (MGT) equations, see e.g. [1, 6, 7, 12, 15,16,17, 19,20,21, 24, 25] and reference therein.
We also observe that it is possible to find in the literature some works dedicated to fourth-order equations in time motivated by MGT equations, see e.g. [1, 7, 16, 19,20,21]; in these papers are made formulations for fourth-order equations of the MGT type, and results of blow-up of solutions, well-posedness, stability and regularity of solutions are obtained.
Our best knowledge indicates that no information is known for models of the type (1.1) concerning the well-posedness and regularity of solutions. This paper makes this point. Furthermore, abstract differential equations of order greater than two are generally ill-posed in the sense of the theory of semigroups of bounded linear operators. However, we identify cases where (1.1) is well-posed. Namely, cases in which this equation is associated with a strongly continuous semigroup and instances in which it is associated with an analytic semigroup.
We will divide our problem into two cases: \(c>0\) and \(c=0\). First of all, note that our linear problem (1.1) can be rewritten as a first-order abstract system of the form
on the phase space
endowed with the usual inner product, where \(U=\left[ {\begin{matrix} u\\ u_t\\ u_{tt}\\ u_{ttt} \end{matrix}}\right] \), and the matrix operator \(S_{a,b,c}\) can be seen as a unbounded linear operator defined by
and
Namely, the unbounded linear operator \(S_{a,b,c}:D(S_{a,b,c})\subset Y\rightarrow Y\) is closed and densely defined.
It is easily seen that \(\lambda \in \rho (-S_{a,b,c})\) if and only if \(\lambda ^4I+a\lambda ^3 A^{\frac{1}{5}}+b\lambda ^2 A^{\frac{4}{5}}+c\lambda A+A^{\frac{2}{5}}\) is bijective. Consequently, if \(\lambda \in \mathbb {C}\) is such that
for some \(\mu \in \sigma (A)\), then \(\lambda \in \sigma (-S_{a,b,c})\). Since \(\sigma (A)\) is countably infinite, we will consider the characteristic polynomials”
with \(\mu _n\in \sigma (A)\). Recall that \(\mu _n>0\) and \(\mu _n\rightarrow \infty \) as \(n\rightarrow \infty \).
Note that the discriminant of (1.2) is given by
see [14, Example on the pages 192-193].
We can summarize the behavior of \(\Delta _n\) when \(n>>0\) as follows (Table 1).
This is a key result in our analysis since a geometric localization of the points of the spectral set of \(-S_{a,b,c}\) can provide us with information about the generation of strongly continuous semigroups of bounded linear operators associated with the problem (1.1), see e.g. [23, Chapter 1, Corollary 3.8].
This work is organized as follows. In Sect. 2 we consider the case \(c>0\) and we treat three subcases \(c=ab\), \(c>ab\), and \(c<ab\). Finally, in Sect. 3 we consider the case \(c=0\) and we treat three subcases \(a=b=c=0\), \(a\ne 0\) and \(b=c=0\), \(b\ne 0\) and \(a=c=0\), and \(a,b\ne 0\) and \(c=0\).
The tables below summarize our results, where \(\checkmark \) means that it is possible to generate the respective strongly continuous semigroups of bounded linear operators on Y, and \(\times \) means that it is not possible to generate the respective strongly continuous semigroups of bounded linear operators on Y.
For \(c>0\), we have the following results (Table 2):
For \(c=0\), we have the following results (Table 3):
2 Case \(c>0\)
Since \(\Delta _n<0\) for large n when \(c>0\) we guarantee the existence of two complex conjugated roots \(z_n\) and \(\overline{z_n}\) and two real roots \(x_n\) and \(y_n\). The following lemma shows the behavior of one of the real roots for the polynomial \(p_n\) in (1.2).
Lemma 1.1
If \(c>0\), then \(y_n\rightarrow 0\) as \(n\rightarrow +\infty \).
Proof
Given \(\epsilon >0\) we have
It is easy to see that \(p_n(-\epsilon )\rightarrow -\infty \) as \(n\rightarrow +\infty \). Since \(p_n(0)>0\) for any n, we guarantee that \(y_n\in (-\epsilon ,0)\) for large n; that is, \(y_n\rightarrow 0\) as \(n\rightarrow +\infty \).
Corollary 1.2
If \((x_n+a\mu _n^{\frac{1}{5}})\rightarrow K\) as \(n\rightarrow +\infty \), then \(Re(z_n)\rightarrow -\dfrac{K}{2}\) as \(n\rightarrow +\infty \).
Proof
It follows directly from Vieta’s formulas
jointly with Lemma 2.1.
Remark 2.3
It will be useful later to know that for any \(\epsilon >0\) we have \(p(-a\mu _n^{\frac{1}{5}}-\epsilon )\) given by
2.1 Subcase \(ab=c\)
In this subcase the polynomial \(p_n\) becomes
We will prove that \(x_n+a\mu _n^{\frac{1}{5}}\rightarrow 0\) and consequently by Corollary 2.2 we have \(Re(z_n)\rightarrow 0\). From this, we will see that it is possible to prove a result of well-posedness for Eq. (1.1).
Proposition 1.4
If \(c>0\) and \(c=ab\), then \(Re(z_n)\rightarrow 0\). This means that, in the best case scenario, the problem (1.1) can only generate a strongly continuous semigroup on Y.
Proof
We will show that \(x_n\rightarrow -a\mu _n^{\frac{1}{5}}\) and the other claim follows from the Hille-Yosida Theorem, see e.g. [23, Chapter 1, Theorem 3.1]. Fixing \(\epsilon >0\) it is easy to see (Remark 2.3) that \(p(-a\mu _n^{\frac{1}{5}}+\epsilon )<0\) for large n. Since \(p_n(-a\mu _n^{\frac{1}{5}})>0\) we know that \(x_n\in (-a\mu _n^{\frac{1}{5}},-a\mu _n^{\frac{1}{5}}+\epsilon )\) for large n; that is, \(x_n\rightarrow -a\mu _n^{\frac{1}{5}}\) as \(n\rightarrow +\infty \).
Theorem 1.5
If \(c>0\) and \(c=ab\), then problem (1.1) is well-posedness on Y; that is, the linear operator associated with this problem is the infinitesimal generator of a strongly continuous semigroup on Y.
Proof
Define
and
Transferring our problem to the phase space
endowed with the usual inner product, we define the vector \(W=\left[ {\begin{matrix} Au\\ Az\\ A^{\frac{3}{5}}z_t\\ A^{\frac{3}{5}}v \end{matrix}}\right] \). It is easy to see that (1.1) can be rewritten as a first-order abstract system of the form
where \(\Lambda \) denotes the unbounded linear operator given by
Namely, this linear operator is closed and densely defined. Moreover, it is a maximally dissipative operator (see [23, Chapter 1, Definition 4.1]) on a space in which the norm is equivalent to the norm of space Z. This also implies that \(\left[ {\begin{matrix} u\\ u_t\\ u_{tt}\\ u_{ttt} \end{matrix}}\right] \in Y\).
2.2 Subcase \(c>ab\)
We will show that in this subcase the problem (1.1) is ill-posed, that is, the operator associated with the problem does not generate a strongly continuous semigroup of bounded linear operators.
Proposition 1.6
If \(c>0\) and \(c>ab\) then \(Re(z_n)\rightarrow +\infty \). Consequently, the unbounded linear operator associated with (1.1) does not generate a strongly continuous semigroup on Y.
Proof
It follows from Remark (2.3) that \(p_n(-a\mu _n^{\frac{1}{5}}-\mu _n^{\frac{1}{10}})<0\) for large n. Since \(p_n(\lambda )\rightarrow +\infty \) as \(\lambda \rightarrow -\infty \) we guarantee that \(x_n<-a\mu _n^{\frac{1}{5}}-\mu _n^{\frac{1}{10}}\). From (2.1) and Lemma 2.1 we obtain \(Re(z_n)\rightarrow +\infty \). This violates a necessary condition for the generation of a strongly continuous semigroup, see e.g. [23, Chapter 1, Corollary 3.8].
2.3 Subcase \(c<ab\)
Note that in this subcase we have to assume that \(a,b,c\ne 0\). We will see that problem (1.1) can be well-posed in the sense of the theory of semigroups of bounded linear operators. However, the semigroup associated with (1.1) cannot be analytic on Y.
Lemma 1.7
If \(a,b,c\ne 0\) and \(c<ab\) then \(x_n\rightarrow \dfrac{-c\mu _n^{\frac{1}{5}}}{b}\) as \(n\rightarrow +\infty \).
Proof
Firstly note that
and therefore \(p_n(-cb^{-1}\mu _n^{\frac{1}{5}})<0\) for large n. Fixing \(\epsilon >0\) we can easily compute that
Hence \(p_n(-cb^{-1}\mu _n^{\frac{1}{5}}-\epsilon )>0\) for large n which guarantees that
for large n; that is, \(x_n\rightarrow -cb^{-1}\mu _n^{\frac{1}{5}}\) as \(n\rightarrow +\infty \).
Corollary 1.8
If \(a,b,c\ne 0\) and \(c<ab\) then \(Re(z_n)\rightarrow (c-ab)2^{-1}b^{-1}\mu _n^{\frac{1}{5}}\). Therefore the problem (1.1) can be well-posed in the sense of the theory of semigroups of bounded linear operators. Moreover, the solution of (1.1) cannot be analytic on Y in the sense of [11].
Proof
From Vieta’s formulas and Lemma 2.7 it follows immediately that \(Re(z_n)\rightarrow (c-ab)2^{-1}b^{-1}\mu _n^{\frac{1}{5}}\) and that
which concludes the proof.
3 Case \(c=0\)
3.1 Subcase \(a=b=c=0\)
In this case, the polynomial is given by
It is easy to see that the roots of \(p_n\) are \(\mu _n^{\frac{1}{10}}e^{\frac{\pi }{4}i}, \mu _n^{\frac{1}{10}}e^{\frac{3\pi }{4}i}, \mu _n^{\frac{1}{10}}e^{\frac{5\pi }{4}i}\) and \(\mu _n^{\frac{1}{10}}e^{\frac{7\pi }{4}i}\). To simplify our view, we can rewrite these roots as
Theorem 1.9
If \(a=b=c=0\), then problem (1.1) is ill-posed in the sense of the theory of semigroups of bounded linear operators on Y.
Proof
Since \(\mu _n\rightarrow +\infty \) our next result follows directly from the Hille-Yosida Theorem, see e.g. [23, Chapter 1, Theorem 3.1].
3.2 Subcase \(a\ne 0\) and \(b=c=0\)
In this case we know that \(\Delta _n<0\) for large n and therefore our polynomial
have two real roots \(x_n\), \(y_n\) and two complex roots \(z_n\) and \(\overline{z}_n\).
Lemma 1.10
If \(a\ne 0\) and \(b=c=0\) then \(x_n\rightarrow -a\mu _n^{\frac{1}{5}}\) as \(n\rightarrow +\infty \).
Proof
Given \(\epsilon >0\) we have
Hence \(p_n(-a\mu _n^{\frac{1}{5}}+\epsilon )<0\) for large n. Since \(p_n(-a\mu _n^{\frac{1}{5}})>0\) for any n, we conclude that \(x_n\in (-a\mu _n^{\frac{1}{5}},-a\mu _n^{\frac{1}{5}}+\epsilon )\) for large n; that is, \(x_n\rightarrow -a\mu _n^{\frac{1}{5}}\) as \(n\rightarrow +\infty \).
Lemma 1.11
If \(a\ne 0\) and \(b=c=0\) then \(y_n\rightarrow - a^{-\frac{1}{3}}\mu _n^{\frac{1}{15}}\) as \(n\rightarrow +\infty \).
Proof
Fix \(\epsilon >0\) and note that
Thus \(p_n(-a^{-\frac{1}{3}}\mu _n^{\frac{1}{15}}-\epsilon )<0\) for large n. Since \(p_n(-a^{-\frac{1}{3}}\mu _n^{\frac{1}{15}})=a^{\frac{-4}{3}}\mu _n^{\frac{4}{15}}\), we guarantee that \(p_n(-a^{-\frac{1}{3}}\mu _n^{\frac{1}{15}})>0\) for large n and therefore
when n is large. This means that \(|y_n+a^{-\frac{1}{3}}\mu _n^{\frac{1}{15}}|\rightarrow 0\) as \(n\rightarrow +\infty \).
Corollary 1.12
If \(a\ne 0\) and \(b=c=0\) then \(Re(z_n)\rightarrow +\infty \). Consequently, problem (1.1) is ill-posed in the sense of the theory of semigroups of bounded linear operators on Y.
3.3 Subscase \(b\ne 0\) and \(a=c=0\)
In this case we have the polynomial
which has \(\Delta _n>0\) for large n and \(P_n=8b\mu _n^{\frac{4}{5}} >0\); that is, our polynomial has two pairs of non-real complex conjugate roots.
Lemma 1.13
If \(b\ne 0\) and \(a=c=0\) then all roots are imaginary for large n. Consequently, problem (1.1) can generate a strongly continuous semigroup of bounded linear operators on Y. Moreover, it cannot generate an analytic semigroup on Y in the sense of [11].
Proof
Denoting the four roots by \(z_n, \overline{z}_n, w_n, \overline{w}_n\) we know from Vieta’s formulas that
and
Assume that \(Re(w_n)\ne 0\). Then
Writing \(w_n\) on its polar form \(w_n=r_ne^{i\theta _n}\) and using (3.1) it follows that
and therefore
From (3.1) we have \(r_n=|w_n|\ne 0\) and this leads to
Case \(sin(4\theta _n)=0\) we have \(sin(2\theta _n)=0\), and \(\theta _n=0\) or \(\theta _n=\frac{\pi }{2}\). From this, \(sin(4\theta _n)\ne 0\); that is,
Putting (3.3) and (3.5) together we obtain
Thus
Since \(b\mu _n^{\frac{3}{5}}+2cos(2\theta _n)\ne 0\) for large n we conclude that \(sin(2\theta _n)=0\) and
which contradicts our assumption that \(w_n\notin \mathbb {R}\) and \(Re(w_n)\ne 0\). Therefore
for large n we conclude the proof.
Theorem 1.14
If \(b\ne 0\) and \(a=c=0\) then the problem (1.1) generates a strongly continuous semigroup on Y.
Proof
It follows similarly to Theorem 2.5.
3.4 Subcase \(a,b\ne 0\) and \(c=0\)
In this case we have the polynomial
which has \(\Delta _n>0\) for large n. Moreover, \(P_n=8b\mu _n^{\frac{4}{5}}-3a^2\mu _n^{\frac{2}{5}}>0\) for large n, which implies that our polynomial has four complex roots \(z_n, \overline{z_n}, w_n\) and \(\overline{w_n}\).
Lemma 1.15
Two of the four roots of polynomial (3.7); namely \(w_n\) and \(\overline{w_n}\) satisfy
Proof
From Vieta’s formulas, we can easily obtain that
and
Since \(w_n,z_n\ne 0\) we can put (3.8) and (3.9) into Eq. (3.10) to obtain that
which leads us to
Now, putting (3.9) and (3.11) together we have
Since \(w_n\ne 0\) we know from (3.13) that \(|w_n|^2-|z_n|^2\ne 0\) and consequently
Putting this equation together with (3.8) we have (the well-defined) equation
Finally, putting (3.12) and (3.14) together we obtain
From this we have
and consequently
Dividing both sides by \(|w_n|^2\) we finally have
Define the polynomial \(q_n:\mathbb {R}\rightarrow \mathbb {R}\) as
From
we obtain that \(q_n(b^{-\frac{1}{2}}\mu _n^{-\frac{1}{5}})<0\) for large n. It is easy to see that for any \(\epsilon >0\) we have \(q_n(b^{-\frac{1}{2}}\mu _n^{-\frac{1}{5}}-\epsilon )>0\) for large n. Thus, given \(\epsilon >0\) we have
for large n, which completes the proof.
Lemma 1.16
Assume that \(w_n\) is a root of polynomial (3.7) such that
as \(n\rightarrow +\infty \). Then \(\dfrac{Re(w_n)}{Im(w_n)}\rightarrow 0\) as \(n\rightarrow +\infty \).
Proof
We will denote the polar form of \(w_n\) as \(r_ne^{i\theta _n}\). Remember that
We just need to prove that \(\theta _n\rightarrow \dfrac{\pi }{2}\) or \(\theta _n\rightarrow \dfrac{3\pi }{2}.\)
Since \(w_n\) is a root of (3.7), we have
It is easy to see that a root from (3.7) must be different from zero. Thus
Since \(-\mu _n^{\frac{2}{5}}r_n^{-2}\rightarrow -b\mu _n^{\frac{4}{5}}\) we conclude that
From the fact that \(r_n\rightarrow b^{-\frac{1}{2}}\mu _n^{-\frac{1}{5}}\) it is easy to see that \(\theta _n\rightarrow \dfrac{\pi }{2}\) or \(\theta _n\rightarrow \dfrac{3\pi }{2}\) as \(n\rightarrow +\infty \).
Corollary 1.17
There exist two roots of polynomial (3.7); namely \(w_n\) and \(\overline{w_n}\) such that \(w_n\rightarrow b^{-\frac{1}{2}}\mu _n^{-\frac{1}{5}}i\) and \(\overline{w_n}\rightarrow -b^{-\frac{1}{2}}\mu _n^{-\frac{1}{5}}i\) as \(n\rightarrow +\infty \).
Proof
It follows immediately from Lemmas 3.7 and 3.8.
Corollary 1.18
If \(a,b\ne 0\) and \(c=0\) then the problem (1.1) can generate a strongly continuous semigroup. Furthermore, the semigroup cannot be of an analytic type on Y in the sense of [11].
Proof
From Corollary 3.9 and Vieta’s formulas, we have that
This means that problem (1.1) can generate a strongly continuous semigroup. Once again from Vieta’s formulas we can obtain that
implying that
Therefore, problem (1.1) cannot have an analytic semigroup on Y as a solution operator.
References
Abouelregal, A.E., Sedighi, H.M., Eremeyev, V.A.: Thermomagnetic behavior of a semiconductor material heated by pulsed excitation based on the fourth-order MGT photothermal model. Continuum Mech. Thermodyn. (2022). https://doi.org/10.1007/s00161-022-01170-z
Amann, H.: Linear and quasilinear parabolic problems. Volume I: Abstract Linear Theory, Birkhäuser Verlag, Basel (1995)
Bezerra, F.D.M., Santos, L.A.: Fractional powers approach of operators for abstract evolution equations of third order in time. J. Differ. Equ. 269(7), 5661–5679 (2020)
Bezerra, F.D.M., Santos, L.A.: Chebyshev polynomials for higher order differential equations and fractional powers. Math. Ann. (2022). https://doi.org/10.1007/s00208-022-02554-x
Bezerra, F. D. M., Santos, L. A.: An extended version of the Cayley–Hamilton–Ziebur Theorem (2022). arxiv:2210.01976
Caixeta, A.H., Lasiecka, I., Cavalcanti, V.N.D.: Global attractors for a third order in time nonlinear dynamics. J. Differ. Equ. 261(1), 113–147 (2016)
Dell’Oro, F., Pata, V.: On a fourth-order equation of Moore–Gibson–Thompson type. Milan J. Math. 85, 215–234 (2017)
Fattorini, H.O.: The cauchy problem. In: Encyclopedia of Mathematics and its Applications, vol. 18. Addison-Wesley Publishing Company, Reading, Massachusetts (1983)
Fattorini, H.O.: Ordinary differential equations in linear topological spaces, I. J. Differ. Equ. 5, 72–105 (1968)
Hasse, M.: The Functional Calculus for Sectorial Operators. Birkhäuser Verlag (2006)
Henry, D.: Geometric theory of semilinear parabolic equations. Lecture Notes in Mathematics, vol. 840. Springer-Verlag, Berlin (1981)
Kaltenbacher, B., Lasiecka, I., Marchand, R.: Well-posedness and exponential decay rates for the Moore–Gibson–Thompson equation arising in high intensity ultrasound. Control Cybernet. 40(4), 971–988 (2011)
Kreǐn, S. G.: Linear differential equations in a banach space. Transl. Mathem. Monogr., 29, American Mathematical Soc. (1971)
Lange, S.: Algebra, Graduate Texts in Mathematics 211, 3rd edn. Springer-Verlag, New York (2002)
Liu, W., Chen, Z., Tiu, Z.: New general decay result for a fourth-order Moore–Gibson–Thompson equation with memory. Electron. Res. Arch. 28(1), 433–457 (2020)
Lizama, C., Murillo, M.: Well-posedness for a fourth-order equation of Moore–Gibson–Thompson type. Electron. J. Qual. Theory Differ. Equ. 81, 1–18 (2021)
Marchand, R., McDevitt, T., Triggiani, R.: An abstract semigroup approach to the third-order Moore–Gibson–Thompson partial differential equation arising in high-intensity ultrasound: structural decomposition, spectral analysis, exponential stability. Math. Methods Appl. Sci. 35(15), 1896–1929 (2012)
Martínez, C., Sanz, M.: Spectral mapping theorem for fractional powers in locally convex spaces, Ann. Scuola Norm. - SCI \(4^e\) série, tome 24(4), 685–702 (1997)
Murillo-Arcila, M.: Well-posedness for the fourth-order Moore–Gibson–Thompson equation in the class of Banach-space-valued Hölder-continuous functions. Math. Methods Appl. Sci. 46(2), 1928–1937 (2023)
Mesloub, A., Zarai, A., Mesloub, F., Cherif, B.B., Abdalla, M.: The Galerkin method for fourth-order equation of the Moore–Gibson–Thompson type with integral condition. Adv. Math. Phys. (2021). https://doi.org/10.1155/2021/5532691
Mesloub, F., Merah, A., Boulaaras, S.: Solution blow-up for a fractional fourth-order equation of Moore–Gibson–Thompson type with nonlinearity nonlocal in time. Math. Notes 113(1), 72–80 (2023)
Neubrander, F.: Well-posedness of higher order abstract Cauchy problems. Trans. Am. Math. Soc. 295, 257–290 (1986)
Pazy, A.: Semigroup of linear operators and applications to partial differential equations. Springer-Verlag, New York (1983)
Pellicer. M., Said-Houari, B.: Wellposedness and decay rates for the cauchy problem of the Moore–Gibson–Thompson equation arising in high intensity ultrasound. Appl. Math. Optim. 1–32 (2017)
Pellicer, M., Solà-Morales, J.: Optimal scalar products in the Moore–Gibson–Thompson equation. Evol. Equ. Control Theory 8(1), 203–220 (2019)
Sobolevskiĭ, P.E.: Equations of parabolic type in a Banach space. Am. Math. Soc. Transl. 49, 1–62 (1966)
Triebel, H.: Interpolation theory, function spaces, differential operators, Veb Deutscher (1978)
Acknowledgements
The authors would like to thank the anonymous referees for their comments and suggestions which greatly improved the work.
Funding
The authors have not disclosed any funding.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
Data sharing does not apply to this article as no datasets were generated or analyzed during the current study. The authors have no relevant financial or non-financial interests to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
F. D. M. Bezerra: Research partially supported by CNPq/Finance Code # 303039/2021-3, Brazil. C. R. Takaessu: Research partially supported by CAPES-PROEX-11169228/D and by FAPESP # 2020/14353-6, Brazil.
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
Bezerra, F.D.M., Santos, L.A., Silva, M.J.M. et al. A Note on the Spectral Analysis of Some Fourth-Order Differential Equations with a Semigroup Approach. Results Math 78, 220 (2023). https://doi.org/10.1007/s00025-023-01999-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00025-023-01999-z
Keywords
- Analytic semigroup
- characteristic polynomial
- fractional power linear problem
- strongly continuous semigroup