K3 surfaces via almost-primes. (Q1599944)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | K3 surfaces via almost-primes. |
scientific article |
Statements
K3 surfaces via almost-primes. (English)
0 references
2002
0 references
This paper shows the importance of the following abstract-looking number-theoretical result in the concrete geometry of \(K3\) surfaces: A deep result due to \textit{H. Iwaniec} [Invent. Math. 47, 171--188 (1978; Zbl 0389.10031)], a first step towards the still open Dirichlet conjecture, asserts that the set of almost-primes (i.e., being a prime, or a square of a prime, or a product of two primes) of the form \(f(n)\) (\(n\) a natural number) is infinite for every fixed irreducible \(f(x) = ax^2 + bx + c\) in \({\mathbb{Z}}[x]\) with \(a > 0\) and \(c\) odd. On the one hand, \textit{D. Orlov} [J. Math. Sci., New York 84, 1361--1381 91997; Zbl 0938.14019)] shows that for complex \(K3\) surfaces \(X_1, X_2\), there exists a Hodge isometry between transcendental lattices \(T_{X_1}\) and \(T_{X_2}\) preserving the line \({\mathbb{C}}\omega_{X_1}\) if and only if the bounded derived categories of coherent sheaves \(D(X_1)\) and \(D(X_2)\) are equivalent as triangulated categories, if and only if \(X_2\) is isomorphic to the base space of a 2-dimensional compact fine moduli space of some stable sheaves on \(X_1\). So a natural question is how much \(D(X)\) or equivalently the pair \((X, \omega_X)\) determines or recovers the surface \(X\). A result of \textit{T. Bridgeland} and \textit{A. Maciocia} [Math. Z. 236, 677--697 (2001; Zbl 1081.14023)] says \(D(X)\) determines \(X\) up to finitely many equivalence classes of non-isomorphic surfaces. In the paper under review, the author successfully solves the remaining problems by proving, among other more general things: 1. For every natural number \(N\), there are at least \(N\) \(K3\) surfaces \(X_i\) with equivalent \(D(X_i)\) but different Picard groups \(\text{Pic}(X_i)\) (and hence non-isomorphic to one another). 2. For a \(K3\) surface \(X\) of Picard number 1, the author calculates precisely, in terms of deg\((X)\), the number of \(K3\) surfaces \(X_i\) with \(D(X_i)\) equivalent to \(D(X)\).
0 references
\(K3\) surfaces
0 references
almost-primes
0 references
transcendental lattice
0 references
NĂ©ron-Severi lattices
0 references