Abstract
We prove the existence and establish the Lifschitz singularity of the integrated density of states for certain random Hamiltonians \(H^\omega =H_0+V^\omega \) on fractal spaces of infinite diameter. The kinetic term \(H_0\) is given by \(\phi (-{\mathcal {L}}),\) where \({\mathcal {L}}\) is the Laplacian on the fractal and \(\phi \) is a completely monotone function satisfying some mild regularity conditions. The random potential \(V^\omega \) is of alloy-type.
Similar content being viewed by others
Availability of data and materials
Not applicable. No datasets were generated or analyzed during the current study.
References
Akkermans, E., Chen, J.P., Dunne, G., Rogers, L.G., Teplyaev, A.: Fractal AC circuits and propagating waves on fractals. Analysis, probability and mathematical physics on fractals, 557–567, Fractals Dyn. Math. Sci. Arts Theory Appl., 5, World Sci. Publ., Hackensack, NJ (2020)
Alonso-Ruiz, P., Baudoin, F., Chen, L., Rogers, L.G., Shanmugalingam, N., Teplyaev, A.: Besov class via heat semigroup on Dirichlet spaces I: Sobolev type inequalities. J. Funct. Anal. 278(11), 108459 (2020)
Alonso-Ruiz, P., Baudoin, F., Chen, L., Rogers, L.G., Shanmugalingam, N., Teplyaev, A.: Besov class via heat semigroup on Dirichlet spaces II: BV functions and Gaussian heat kernel estimates. Calc. Var. Partial Differ. Equ. 59(3), 103 (2020)
Alonso-Ruiz, P., Baudoin, F., Chen, L., Rogers, L.G., Shanmugalingam, N., Teplyaev, A.: Besov class via heat semigroup on Dirichlet spaces III: BV functions and sub-Gaussian heat kernel estimates. Calc. Var. Partial Differ. Equ. 60(5), 170 (2021)
Alonso-Ruiz, P., Baudoin, F., Chen, L., Rogers, L.G., Shanmugalingam, N., Teplyaev, A.: BV functions and fractional Laplacians on Dirichlet spaces. arXiv:1910.13330
Balsam, H., Kaleta, K., Olszewski, M., Pietruska-Pałuba, K.: IDS for subordinate Brownian motions in Poisson random environment on nested fractals: existence and Lifschitz tail. preprint (2022)
Benderskii, M., Pastur, L.: On the spectrum of the one-dimensional Schrödinger equation with random potential. Mat. Sb. 82, 245–256 (1970)
Bourgain, J., Kenig, C.E.: On localization in the continuous Anderson–Bernoulli model in higher dimension Invent. Invent. Math. 161, 389–426 (2005)
Carmona, R., Lacroix, J.: Spectral Theory of Random Schrödinger Operators. Probability and Its Applications. Birkhäuser Boston Inc., Boston (1990)
Chaumont, L., Uribe Bravo, G.: Markovian bridges: weak continuity and pathwise constructions. Ann. Probab. 39(2), 609–647 (2011)
Chen, J.P., Hinz, M., Teplyaev, A.: From non-symmetric particle systems to non-linear PDEs on fractals. Stochastic partial differential equations and related fields, 503–513, Springer Proc. Math. Stat., 229, Springer, Cham (2018)
Chen, Z.-Q., Song, R.: Two sided eigenvalue estimates for subordinate processes in domains. J. Funct. Anal. 226, 90–113 (2005)
Chung, K.L., Zhao, Z.: From Brownian Motion to Schrödinger’s Equation. Springer, New York (1995)
Combes, J.M., Hislop, P.D.: Localization for some continuous, random Hamiltonians in d-dimensions. J. Funct. Anal. 124, 149–180 (1994)
Dekkers, A., Rozanova-Pierrat, A., Teplyaev, A.: Mixed boundary valued problems for linear and nonlinear wave equations in domains with fractal boundaries. Calc. Var. Partial Differ. Equ. 61(2), 75 (2022)
Demuth, M., van Casteren, J.A.: Stochastic Spectral Theory for Self-adjoint Feller Operators. A Functional Analysis Approach. Birkhäuser, Basel (2000)
Friedberg, R., Luttinger, J.: Density of electronic energy levels in disordered systems. Phys. Rev. B 12, 4460–4474 (1975)
Fukushima, M.: Dirichlet forms, diffusion processes, and spectral dimensions for nested fractals. In: Ideas and Methods in Stochastic Analysis. Stochastics and Applications, pp. 151–161. Cambridge University Press, Cambridge (1992)
Fitzsimmons, P.J., Hambly, B.M., Kumagai, T.: Transition density estimates for Brownian motion on affine nested fractals. Commun. Math. Phys. 165(3), 595–620 (1994)
Fukushima, M.: On the spectral distribution of a disordered system and a range of a random walk. Osaka J. Math. 11, 73–85 (1974)
Fukushima, M., Nagai, H., Nakao, S.: On an asymptotic property of spectra of a random difference operator. Proc. Jpn. Acad. 51, 100–102 (1975)
Gebert, M., Rojas-Molina, C.: Lifshitz tails for the fractional Anderson model. J. Stat. Phys. 179, 341–353 (2020)
Germinet, F., Hislop, P.D., Klein, A.: Localization for Schrödinger operators with Poisson random potential. J. Eur. Math. Soc. (JEMS) 9(3), 577–607 (2007)
Hinz, M., Rozanova-Pierrat, A., Teplyaev, A.: Non-Lipschitz uniform domain shape optimization in linear acoustics. SIAM J. Control Optim. 59(2), 1007–1032 (2021)
Kumagai, T., Nakamura, C.: Lamplighter random walks on fractals. J. Theor. Probab. 31(1), 68–92 (2018)
Hinz, M., Teplyaev, A.: Dirac and magnetic Schrödinger operators on fractals. J. Funct. Anal. 265(11), 2830–2854 (2013)
Kaleta, K., Olszewski, M., Pietruska-Pałuba, K.: Reflected Brownian motion on simple nested fractals. Fractals 27(6), 19501041–29 (2019)
Kaleta, K., Pietruska-Pałuba, K.: Integrated density of states for Poisson–Schrödinger perturbations of subordinate Brownian motions on the Sierpiński gasket. Stoch. Process. Appl. 125(4), 1244–1281 (2015)
Kaleta, K., Pietruska-Pałuba, K.: Lifschitz singularity for subordinate Brownian motions in presence of the Poissonian potential on the Sierpiński triangle. Stoch. Process. Appl. 128(11), 3897–3939 (2018)
Kaleta, K., Pietruska-Pałuba, K.: Lifschitz tail for alloy-type models driven by the fractional Laplacian. J. Funct. Anal. 279(5), 108575 (2020)
Kaleta, K., Pietruska-Pałuba, K.: Lifschitz tail for continuous Anderson models driven by Lévy operators. Commun. Contemp. Math. 23(6), 2050065 (2021)
Kigami, J.: Time changes of the Brownian motion: poincaré inequality, heat kernel estimate and protodistance. Mem. Am. Math. Soc. 259(1250), v+118 (2019)
Kirsch, W., Martinelli, F.: On the density of states of Schrödinger operators with a random potential. J. Phys. A 15, 2139–2156 (1982)
Kirsch, W., Martinelli, F.: Large deviations and Lifshitz singularity of the integrated density of states of random Hamiltonians. Commun. Math. Phys. 89, 27–40 (1983)
Kirsch, W., Simon, B.: Lifshitz tails for periodic plus random potentials. J. Stat. Phys. 42(5/6), 799–808 (1986)
Kirsch, W., Veselić, I.: Lifshitz tails for a class of Schrödinger operators with random breather-type potential. Lett. Math. Phys. 94, 27–39 (2010)
Klopp, F.: Weak disorder localization and Lifshitz tails. Commun. Math. Phys. 232, 125–155 (2002)
Klopp, F.: Weak disorder localization and Lifshitz tails: continuous Hamiltonians. Ann. Henri Poincaré 3, 711–737 (2002)
Kumagai, T.: Estimates of transition densities for Brownian motion on nested fractals. Probab. Theory Relat. Fields 96(2), 205–224 (1993)
Kumagai, T.: Anomalous random walks and diffusions: from fractals to random media. In: Proceedings of the International Congress of Mathematicians, Seoul 2014. Vol. IV, 75–94, Kyung Moon Sa, Seoul (2014)
Kusuoka, S.: Dirichlet forms on fractals and products of random matrices. Publ. RIMS Kyoto Univ. 25, 659–680 (1989)
Lindstrom, T.: Brownian motion on nested fractals. Mem. Am. Math. Soc. 83(420), iv+128 (1990)
Luttinger, J.M.: New variational method with applications to disordered systems. Phys. Rev. Lett. 37, 609–612 (1976)
Mezincescu, G.: Bounds on the integrated density of electronic states for disordered Hamiltonians. Phys. Rev. B 32, 6272–6277 (1985)
Nagai, H.: On an exponential character of the spectral distribution function of a random difference operator. Osaka J. Math. 14, 111–116 (1977)
Nakao, S.: On the spectral distribution of the Schrödinger operator with random potential. Jpn. J. Math. 3, 111–139 (1977)
Nieradko, M., Olszewski, M.: Good labeling property of simple nested fractals (2021) arXiv:2110.15921 to appear in Journal of Fractal Geometry
Okura, H.: On the spectral distributions of certain integro-differential operators with random potential. Osaka J. Math. 16(3), 633–666 (1979)
Olszewski, M.: Estimates for the transition densities of the reflected Brownian motion on simple nested fractals. Prob. Math. Statist. 39(2), 423–440 (2019)
Pastur, L.A.: The behavior of certain Wiener integrals as \(t\rightarrow \infty \) and the density of states of Schrödinger equations with random potential. Teoret. Mat. Fiz. 32, 88–95 (1977). (Russian)
Pietruska-Pałuba, K.: The Lifschitz singularity for the density of states on the Sierpinski gasket. Probab. Theory Relat. Fields 89(1), 1–33 (1991)
Pietruska-Pałuba, K.: The wiener sausage asymptotics on simple nested fractals. Stoch. Anal. Appl. 23(1), 111–135 (2005)
Pietruska-Pałuba, K., Stos, A.: Poincaré inequality and Hajlasz–Sobolev spaces on nested fractals. Studia Math. 218(1), 1–28 (2013)
Romerio, M., Wreszinski, W.: On the Lifshitz singularity and the tailing in the density of states for random lattice systems. J. Stat. Phys. 21, 169–179 (1979)
Reed, M., Simon, B.: Methods on Modern Mathematical Physics. Vol. 4: Analysis of Operators. Academic Press, Cambridge (1978)
Schilling, R., Song, R., Vondraček, Z.: Bernstein Functions. Walter de Gruyter, Berlin (2010)
Shima, T.: Lifschitz tails for random Schrödinger operators on nested fractals. Osaka J. Math. 29, 749–770 (1992)
Simon, B.: Lifshitz tails for the Anderson model. J. Stat. Phys. 38, 65–76 (1985)
Stollmann, P.: Caught by Disorder: Bound States in Random Media. Birkhäuser, Boston (2001)
Sznitman, A.S.: Lifschitz tail and Wiener sausage on hyperbolic space. Commun. Pure Appl. Math. 42, 1033–1065 (1989)
Sznitman, A.S.: Lifschitz tail on hyperbolic space: Neumann conditions. Commun. Pure Appl. Math. 43, 1–30 (1990)
Acknowledgements
We thank the referee for a careful reading of the paper and useful comments.
Funding
Research was supported by the National Science Centre, Poland, Grants No. 2015/17/B/ST1/01233 and 2019/35/B/ST1/02421.
Author information
Authors and Affiliations
Contributions
All the authors equally contributed to this article.
Corresponding author
Ethics declarations
Conflict of interest
The authors have no competing interests to declare that are relevant to the content of this article.
Ethical approval
Not applicable.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Research was supported in parts by the National Science Centre, Poland, Grants No. 2015/17/B/ST1/01233 and 2019/35/B/ST1/02421.
Appendices
Appendix A: Technical lemmas on subordinate processes
We need the following technical results.
Lemma A.1
Let \({\mathcal {K}}^{\langle \infty \rangle }\) be an USNF with the GLP and let the assumption (B) hold. For every \(t>0\) and \(a>1\) we have
Proof
The proof follows the lines of that of [28, Lemma 2.3]. The only difference is that now we use the Euclidean distance instead of the geodesic one, and apply the sub-Gaussian estimates from [19, Lemma 5.6, Remark 3.7]. At the end, we also use the summation property (2.20) as before. \(\square \)
Let us recall that \(\Delta _{M,i}\) for \(1 \le i \le N\) denote the M-complexes in \({\mathcal {K}}^{\langle M+1 \rangle }\), see Definition 2.1 (9).
Lemma A.2
Let \({\mathcal {K}}^{\langle \infty \rangle }\) be an USNF with the GLP and let the assumption (B) hold. We have the following.
-
(a)
For
$$\begin{aligned} C(M,t):= \sup _{x,y \in {\mathcal {K}}^{\langle M \rangle }} \sum _{\begin{array}{c} y^{\prime } \in \pi _{M}^{-1}(y) \\ y' \notin {\mathcal {K}}^{\left\langle M+1 \right\rangle } \end{array}} p(t,x,y'), \quad t>0, \quad M \in {\mathbb {Z}} \end{aligned}$$(A.1)we have
$$\begin{aligned} \sum _{M=1}^{\infty } C(M,t) < \infty , \quad t>0; \end{aligned}$$in particular, for every \(t>0\), \(C(M,t) \rightarrow 0\) as \(M \rightarrow \infty \).
-
(b)
For every \(t>0\),
$$\begin{aligned} \sum _{M=1}^{\infty } \frac{1}{\mu \left( {\mathcal {K}}^{\langle M \rangle }\right) } \int _{{\mathcal {K}}^{\langle M \rangle }} \left| p(t,x,x) - p_M(t,x,x) \right| \mu (\textrm{d}x) < \infty ; \end{aligned}$$(A.2)in particular
$$\begin{aligned} \frac{1}{\mu \left( {\mathcal {K}}^{\langle M \rangle }\right) } \int _{{\mathcal {K}}^{\langle M \rangle }} \left| p(t,x,x) - p_M(t,x,x) \right| \mu (\textrm{d}x) \rightarrow 0, \quad as \ M \rightarrow \infty .\nonumber \\ \end{aligned}$$(A.3)
Proof
By Tonelli’s theorem and [49, formula (3.3) and Lemma 3.5], we get
Further steps of the proof follow exactly the reasoning in the proof of [28, Lemma 2.5 (b)].
The proof of (b) is similar to that of [28, Lemma 2.7], i.e. we start with
and first observe that \({\mathcal {B}}_M(t)\) is a term of convergent series by part (a). To prove that the same is true for \({\mathcal {A}}_M(t)\), we only need to modify the estimate in the proof of the quoted lemma. The difference is that the domain of the integration in \({\mathcal {A}}_M(t)\) has to be divided into two different sets \(E_M^1\) and \(E_M^2\). In the present general case, we cannot use geodesic balls. We will apply the graph distance here.
Let us denote \(V_M^{\langle M \rangle } = \{v_1,\ldots , v_k\}\) and
Recall that \(\sup _{(x,y) \in {\mathcal {K}}^{\langle \infty \rangle } \times {\mathcal {K}}^{\langle \infty \rangle }} p(t,x,y) < \infty \) by Lemma 2.1 (a). Since \(E_M^2\) consists of k \(\lfloor M/2 \rfloor \)-complexes (each one attached to one of the vertices from \(V_M^{\langle M \rangle }\)), we have that
In consequence, the part of \({\mathcal {A}}_M(t)\) including the integral over \(E_M^2\) is a term of a convergent series.
When \(x \in E_M^1\) and \(x^{\prime } \in \pi _{M}^{-1}(x) \backslash {\mathcal {K}}^{\langle M \rangle }\), then \(d_{\lfloor M/2 \rfloor } (x, x') >2\), so from [27, Lemma A.2] we have \(|x-x'|> c_4 L^{M/2}\). Using the subordination formula for the density p(t, x, y) and the subgaussian estimates from [39] for the density g(u, x, y), we then get
Collecting these estimates, we obtain
with the constants \(c_6, c_7\) independent of M, and from this point the proof can be continued in the same manner as that of [28, Lemma 2.7]. \(\square \)
Appendix B: Proofs of the statements from Sect. 3
Proof of Proposition 3.1
Fix \(t>0\). For \(\mu \)-almost all \(x \in {\mathcal {K}}^{\langle M+1\rangle },\) by Lemmas 2.2(a), 3.1, and the inclusion \(\pi _{M+1}^{-1}(x) \subset \pi _M^{-1}\left( \pi _M(x)\right) \), we may write
Using this estimate, the definition of \(\Lambda _{M^*}^{N}(t,\omega )\) and Lemma 2.2(b), we may now follow the argument in the second part of the proof of [28, Theorem 3.1], getting that
Since \({\mathbb {E}}_{{\mathbb {Q}}}\Lambda _{M}^{N, V_M^\omega }(t) \ge 0\), it converges to a finite limit \(\Lambda (t)\) as \(M \rightarrow \infty \). This completes the proof. \(\square \)
Proofs of Lemmas 3.2 and 3.3 given below are of technical nature. They follow the steps and ideas from the proofs of Proposition 3.1 and Lemmas 3.1\(-\)3.2 in [28]. The main difference here is that we now work with different type of random potentials and the state space is now a general USNF (let us emphasize that the Sierpiński triangle which was studied in the quoted paper is one of the simplest planar nested fractals with high regularity). This causes some extra geometric issues, which are solved by using the graph distance (the geodesic metric may not be defined at all) and the comparison principle from [27, Lemma A.2]. Therefore, we will follow only the main steps of the proofs which are affected by these changes, focusing on the most critical differences.
Proof of Lemma 3.2
We first show (a). By following the argument at the beginning of [28, Proposition 3.1] we obtain that for each fixed \(t>0\) there exists a constant \(c=c(t)\) such that
where
Note that the above bound does not depend on \(\omega \). By Lemma A.2 (b), \(R_{2,M}(t)\) is the term of a convergent series, so we only need to estimate \(R_{1,M}(t)\).
Denote the vertices from \({\mathcal {V}}_M^{\langle M \rangle }\) as \(v_i\), \(1\le i\le k\), and let \(\Delta _{\left\lfloor M/2 \right\rfloor ,v_i} \subset {\mathcal {K}}^{\langle M \rangle }\), be the \(\left\lfloor M/2 \right\rfloor \)-complex attached to \(v_i\). If the process starts from \(x \in {\mathcal {D}}_M:={\mathcal {K}}^{\langle M \rangle } \backslash \bigcup _{i=1}^{k} \Delta _{\left\lfloor M/2 \right\rfloor ,v_i}\), then, using [27, Lemma A.2], we have
with a constant \(c_1\), independent of M. We also have
Recall that \(\mu \left( {\mathcal {K}}^{\langle M \rangle } \backslash {\mathcal {D}}_M \right) = k N^{\left\lfloor M/2 \right\rfloor }\). Then, by using the upper bound in Lemma 2.1(a), the formula (2.25) and the symmetry of the bridge measure, we get
Therefore
which is, by Lemma A.1, a term of a convergent series. The proof of (a) is completed.
The proof of the first convegence in (b) follows the lines of that of (a) and it is omitted. We only show the second convergence.
First observe that
where
By using exactly the same notation and argument as in the proof of part (a), we may write
We see that the first term converges to 0 as \(M \rightarrow \infty \). It is sufficient to show that the second term goes to zero as well; denote it by I(t, M). We have
To estimate \(I_1(t,M)\) we will use the inequality \(|\textrm{e}^{-x}-\textrm{e}^{-y}| \le |x-y|, x,y>0\), the fact that W and \(\xi _y\) are nonnegative.
Now, observe that for every \(x \in {\mathcal {K}}^{\langle \infty \rangle }\) we have
and
By this observation, the Fubini theorem and the fact that all the lattice random variables together with the profile function W are nonnegative, we get that the above expectation can be estimated above by
As \(X_s \in {\mathcal {D}}_M\) for \(0\le s \le t\) and \(y \in {\mathcal {V}}_{0}^{\langle \infty \rangle } \backslash {\mathcal {V}}_{0}^{\langle M \rangle }\), we have \(d_{\left\lfloor M/2 \right\rfloor }(X_s, y)>2\) for \(0\le s\le t\). This gives
with a constant \(c_5>0\) independent of M. By (W3) this is a term of a convergent series.
To show that \(I_2(t,M)\) is a term of a convergent series, we just follow the steps from the proof of (a) for an appropriate set \({\mathcal {D}}^{'}_M \subset {\mathcal {D}}_M\). The proof of (b) is finished. \(\square \)
Proof of Lemma 3.3
The proof of this lemma follows the main steps from the proof of [28, Lemma 3.2], but it substantially differs in some details which are critical for the argument. We will focus on explanation of these differences.
First note that due to Lemma 3.2(a), similarly as in the proof of [28, Lemma 3.2], we only need to establish (3.2). We consider a family of measures \((\nu _M)_{M \in \mathbb {Z_+}}\) given by
defined with the product space \({\widetilde{\Omega }} = D([0,t],{\mathcal {K}}^{\langle \infty \rangle })^2 \times \Omega ^3\), and nondecreasing sequence of positive integers \(\left( a_M\right) _{M \in {\mathbb {Z}}_{+}}\) such that \(a_M \le M\), \(M \in {\mathbb {Z}}_{+}\). Its values will be chosen later in the proof.
For \(m \in {\mathbb {Z}}_{+}\) we set
and
where the ball \(B_m(x,1)\) is taken in the m-graph metric, i.e. for \(x \in {\mathcal {K}}^{\langle \infty \rangle } \backslash {\mathcal {V}}^{\langle \infty \rangle }_m\) it is equal to \(\Delta _m(x)\) – the only m-complex containing x; for \(x \in {\mathcal {V}}^{\langle \infty \rangle }_m\) it is a sum of m-complexes attached to x (there are \(\text {rank}(x) \in \{1,2,3\}\) of them). We also denote for \(M \in {\mathbb {Z}}_{+}\)
With this notation we have
We now consider the partition of \({\widetilde{\Omega }}\) into three disjoint sets:
We will integrate \({\mathcal {X}}\) over each of these sets separately. Let us point out that all these sets are now defined with the m-graph metric \(d_m(x,y)\). We also use this opportunity to correct the definition of the sets \(D_0^M\) and \(D_1^M\) on p. 1272 in [28]: they also should be defined with \(s_1,s_2 \in [0,t]\) as above instead of single \(s \in [0,t]\). Also, \(a_M\) should be used in the definition of \(D_0^M\) instead of \(c_M\).
By following the argument in the proof of the quoted lemma, we get
For a given \(M \in {\mathbb {Z}}_{+}\) and a path \(X_s(w)\), the functional \(F_M(w,\cdot )\) depends only on those fractal lattice points from \(V_{0}^{\langle \infty \rangle }\) which are in the set \(X_{[0,t]}^{a_M}(w):= \bigcup _{s\in [0,t]} B_{a_M}(X_s(w),1)\). From the definition of \(D_0^M\) we see that on this set we have \(X_{[0,t]}^{a_M}(w_1) \cap X_{[0,t]}^{a_M}(w_2) = \emptyset \) and therefore, the random variables \(F_M(w_1,\omega _0) - F_M(w_1,\omega _1)\) and \(F_M(w_2,\omega _0) - F_M(w_2,\omega _2)\) are \({\mathbb {Q}}^{\otimes 3}\)-independent. In consequence,
and, by following the argument in the proof of the cited lemma, including Jensen’s inequality and the assumption that all lattice random variables are nonnegative and integrable, we obtain
for a constant \(c>0\), independent of M.
On the set \(D_1^M\) we have
This can be seen as follows (see Fig. 1 for illustration). From the definition of \(D_1^M\) we have that \(d_{a_M+3} \left( X_0(w_1), X_0(w_2) \right) >2\), what means that \(X_0(w_1)\) and \(X_0(w_2)\) are in separate \((a_M+3)\)-complexes (light grey complexes in the figure). If, on the contrary to (B.5), for both \(i \in \{1,2\}\) and all \(s \in (0,t]\) were \(d_{a_M} \left( X_0(w_i), X_s(w_i) \right) \le 2\), then for both \(i=1,2\) the entire path \(X_{s}(w_i)\) would be bounded inside \(B_{a_M}(X_0(w_i),2)\) (dark grey complexes in the figure). That would mean that these two paths of the process up to time t are not closer to each other than in two different \(a_M\)-complexes attached to the vertices of a common \((a_M+3)\)-complex (white in the figure). In fact, they would be in separate \((a_M+1)\)-complexes inside two different \((a_M+2)\)-complexes. This would mean that \(d_{a_M} \left( X_{s_1}(w_1), X_{s_2}(w_2) \right) \ge 6\) for all \(s_1, s_2 \in (0,t]\), because the path realizing the graph distance must pass through another \(a_M\)-complex inside \((a_M+1)\)-complex containing \(X_{s_1}(w_1)\), then through another \((a_M+1)\)-complex inside \((a_M+2)\)-complex containing \(X_{s_1}(w_1)\), so at least two more \(a_M\)-complexes. By symmetry, it must go next through at least three \(a_M\)-complexes inside \((a_M+2)\)-complex containing \(X_{s_2}(w_2)\).
By [27, Lemma A.2] this implies that
with a constant \(c_1\) independent of M, and since \(0\le F_M(w_i,\omega _k) \le 1\), \(i=1,2\), \(k=0,1,2\), the integral over \(D_1^M\) can be estimated by
Using the same argument as in the same step of the proof of the cited lemma, we then get that the above expression is less than or equal to
with some \(c_2\), independent of M.
Since the integrand \({\mathcal {X}}\) is not bigger than 1, it is enough to estimate the measure of \(D_2^M\). We have
with some \(c_3\) independent of M.
We may choose \(a_M=\lfloor M/4 \rfloor \). Then (B.4) is a term of a convergent series by (W3) and (B.7) is a term of convergent series by Lemma A.1. \(\square \)
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
Balsam, H., Kaleta, K., Olszewski, M. et al. Density of states for the Anderson model on nested fractals. Anal.Math.Phys. 14, 23 (2024). https://doi.org/10.1007/s13324-024-00880-8
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s13324-024-00880-8
Keywords
- Integrated density of states
- Alloy-type potential
- Subordinate Brownian motion
- Nested fractal
- Good labelling property
- Reflected process
- Neumann boundary conditions