Skip to main content
Log in

\({ TMF }_0(3)\)-Characteristic classes for string bundles

  • Published:
Mathematische Zeitschrift Aims and scope Submit manuscript

Abstract

We compute the completed \({ TMF }_0(3)\)-cohomology of the 7-connected cover \( BString \) of \( BO \). We use cubical structures on line bundles over elliptic curves to construct an explicit class which together with the Pontryagin classes freely generates the cohomology ring.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Anderson, D.W., Brown Jr, E.H., Peterson, F.P.: The structure of the spin cobordism ring. Ann. Math. (2) 86, 271–298 (1967)

    Article  MATH  MathSciNet  Google Scholar 

  2. Ando, M., Hopkins, M.J., Rezk, C.: Multiplicative orientations of KO-theory and of the spectrum of topological modular forms. Unpublished (2010)

  3. Ando, M., Hopkins, M.J., Strickland, N.P.: Elliptic spectra, the Witten genus and the theorem of the cube. Invent. Math. 146(3), 595–687 (2001)

    Article  MATH  MathSciNet  Google Scholar 

  4. Atiyah, M.F.: \(K\)-theory and reality. Q. J. Math. Oxf. Ser. (2) 17, 367–386 (1966)

    Article  MATH  MathSciNet  Google Scholar 

  5. Deligne, P., Rapoport, M.: Les schémas de modules de courbes elliptiques, Modular functions of one variable, II (Proc. Internat. Summer School, Univ. Antwerp, Antwerp, 1972). Lecture Notes in Mathematics, Vol. 349, pp. 143–316. Springer, Berlin (1973)

  6. Dugger, D.: An Atiyah-Hirzebruch spectral sequence for \(\mathit{KR}\)-theory. K-Theory 35(3–4), 213–256 (2006)

    Article  MathSciNet  Google Scholar 

  7. Elmendorf, A.D., Kriz, I., Mandell, M.A., May, J.P.: Rings, modules, and algebras in stable homotopy theory. In: Mathematical Surveys and Monographs, vol. 47. American Mathematical Society, Providence, RI (With an appendix by M. Cole) (1997)

  8. Goerss, P.G.: Topological modular forms [after Hopkins, Miller and Lurie], Astérisque (2010), no. 332, Exp. No. 1005, viii, 221–255, Séminaire Bourbaki. Volume 2008/2009. Exposés 997-1011

  9. Hirzebruch, F., Berger, T., Jung, R.: Manifolds and modular forms, Aspects of mathematics, E20, Friedr. Vieweg & Sohn, Braunschweig (With appendices by Nils-Peter Skoruppa and by Paul Baum) (1992)

  10. Hu, P., Kriz, I.: Real-oriented homotopy theory and an analogue of the Adams-Novikov spectral sequence. Topology 40(2), 317–399 (2001)

    Article  MATH  MathSciNet  Google Scholar 

  11. Hopkins, M.J.: Algebraic topology and modular forms. In: Proceedings of the International Congress of Mathematicians, vol. I, pp. 291–317. Higher Education Press, Beijing (2002)

  12. Hovey, M., Strickland, N.P.: Morava \(K\)-theories and localisation. Mem. Am. Math. Soc. 139(666), viii+100 (1999)

    MathSciNet  Google Scholar 

  13. Hu, P.: The \({{\rm Ext}}^0\)-term of the real-oriented Adams–Novikov spectral sequence. In: Homotopy Methods in Algebraic Topology (Boulder, CO, 1999), Contemporary Mathematics, vol. 271, pp. 141–153. American Mathematical Society, Providence, RI (2001)

  14. Kitchloo, N., Laures, G.: Real structures and Morava \(K\)-theories. K-Theory 25(3), 201–214 (2002)

    Article  MATH  MathSciNet  Google Scholar 

  15. Kitchloo, N., Laures, G., Wilson, W.S.: The Morava \(K\)-theory of spaces related to \(\mathit{BO}\). Adv. Math. 189(1), 192–236 (2004)

    Article  MATH  MathSciNet  Google Scholar 

  16. Kronholm, W.C.: The \({{\rm RO}}(G)\)-graded Serre spectral sequence. Homol. Homotopy Appl. 12(1), 75–92 (2010)

    Article  MATH  MathSciNet  Google Scholar 

  17. Kitchloo, N., Wilson, W.S.: Unstable splittings for real spectra. Algebr. Geom. Topol. 13(2), 1053–1070 (2013)

    Article  MATH  MathSciNet  Google Scholar 

  18. Kitchloo, N., Wilson, W.S.: The \(\mathit{ER(n)}\)-cohomology of \(\mathit{BO(q)}\), and real Johnson-Wilson orientations for vector bundles. Bull. Lond. Math. Soc. 47, 835–847 (2015)

    Article  MathSciNet  Google Scholar 

  19. Laures, G.: Characteristic classes in \(T\!M\!F\) of level \(\ \Gamma _1(3)\). Trans. AMS. (2014, to appear). arXiv:1304.3588

  20. Laures, G.: \(K(1)\)-local topological modular forms. Invent. Math. 157(2), 371–403 (2004)

    Article  MATH  MathSciNet  Google Scholar 

  21. May, J.P.: Equivariant homotopy and cohomology theory, CBMS Regional Conference Series in Mathematics, vol. 91, Published for the Conference Board of the Mathematical Sciences, Washington, DC; by the American Mathematical Society, Providence, RI, 1996, With contributions by M. Cole, G. Comezaña, S. Costenoble, A. D. Elmendorf, J. P. C. Greenlees, L. G. Lewis, Jr., R. J. Piacenza, G. Triantafillou, and S. Waner

  22. Miller, H.: The elliptic character and the Witten genus. In: Algebraic topology (Evanston, IL, 1988), Contemporary Mathematics, vol. 96, pp. 281–289. American Mathematical Society, Providence, RI (1989)

  23. Mahowald, M., Rezk, C.: Topological modular forms of level 3. Pure Appl. Math. Q. 5(2) (2009) (Special Issue: In honor of Friedrich Hirzebruch. Part 1, 853–872)

  24. Novikov, S.P.: Homotopy properties of Thom complexes. Mat. Sb. (N.S.) 57(99), 407–442 (1962)

    MathSciNet  Google Scholar 

  25. Ravenel, D.C.: Localization with respect to certain periodic homology theories. Am. J. Math. 106(2), 351–414 (1984)

    Article  MATH  MathSciNet  Google Scholar 

  26. Ravenel, D.C., Wilson, W.S.: The Morava \(K\)-theories of Eilenberg–Mac Lane spaces and the Conner–Floyd conjecture. Am. J. Math. 102(4), 691–748 (1980)

    Article  MATH  MathSciNet  Google Scholar 

  27. Ravenel, D.C., Wilson, W.S.: The Hopf ring for complex cobordism. J. Pure Appl. Algebra 9(3), 241–280 (1976/1977)

  28. Ravenel, D.C., Wilson, W.S., Yagita, N.: Brown-Peterson cohomology from Morava \(K\)-theory. K-Theory 15(2), 147–199 (1998)

    Article  MATH  MathSciNet  Google Scholar 

  29. Silverman, J.H.: The arithmetic of elliptic curves. Graduate Texts in Mathematics, 2nd edn, vol. 106. Springer, Dordrecht (2009)

  30. Su, H.-H.: The \(E(1,2)\) cohomology of the Eilenberg-MacLane space \(K(\mathbb{Z},3)\), ProQuest LLC. Ph.D. Thesis, The Johns Hopkins University, Ann Arbor, MI (2007)

  31. Thom, R.: Quelques propriétés globales des variétés différentiables. Comment. Math. Helv. 28, 17–86 (1954)

    Article  MATH  MathSciNet  Google Scholar 

Download references

Acknowledgments

The authors would like to thank Nitu Kitchloo, Björn Schuster and Vesna Stojanoska for helpful discussions. They are also grateful to the referee for a very careful revision.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Gerd Laures.

Appendices

Appendix 1: The invariance of the Pontryagin classes and the difference class

We would like to give a geometric proof for the invariance of the Pontryagin classes and the difference class.

An isomorphism of Weierstrass curves over a ring R with projective coordinates [X : Y : Z] and \([X':Y':Z']\) respectively is in general given by

$$\begin{aligned} X'=u^2X+rZ,\quad Y'=su^2X+u^3Y+tZ,\quad Z'=Z, \end{aligned}$$

where \(u\in R^\times , r,s,t\in R\). We denote by x respectively \(x'\) the coordinates on the corresponding formal groups given by the restrictions of the functions \(\frac{X}{Y}\) and \(\frac{X'}{Y'}\) respectively. On the formal group, the function \(z=\frac{Z}{Y}\) becomes a power series in the coordinate x. This implies

$$\begin{aligned} x'=\frac{u^{-1}x+ru^{-3}z}{1+tu^{-3}z+su^{-1}x}=g(x), \end{aligned}$$

where g is a power series in one variable with vanishing constant term and invertible linear coefficient. It is an isomorphism of the formal group laws induced by the coordinates x and \(x'\) respectively.

In particular we are interested in the universal triple \((C,\omega ,P)\) of an elliptic curve with invariant differential and level structure consisting of a point P of order 3. This is

$$\begin{aligned} C\,{:}\,Y^2Z+a_1XYZ+a_3YZ^2=X^3 \end{aligned}$$

over \(E^*={ TMF }_1(3)^*={\mathbb Z}_{(2)}[a_1,a_3,\Delta ^{-1}]\) with \(\omega =\frac{dX}{2Y+a_1X+a_3}\) and \(P=(0,0)\).

As \((C,\omega ,P)\) is the universal triple and \((C,\omega ,-P)\) is another such triple, there exists a ring endomorphism of \({ TMF }_1(3)^*={\mathbb Z}_{(2)}[a_1,a_3,\Delta ^{-1}]\) such that the pushforward of \((C,\omega ,P)\) is isomorphic to \((C,\omega ,-P)\). In fact this endomorphism is an automorphism of order 2: it sends \(a_1\mapsto -a_1,a_3\mapsto -a_3\).

The reason is that \((C,\omega ,-P)\) is isomorphic to \((C^-,\omega ',P=(0,0))\), where

$$\begin{aligned} C^-&:\quad (Y')^2Z'-a_1X'Y'Z'-a_3Y'(Z')^2=(X')^3,\\ \omega '&=\frac{dX'}{2Y'-a_1X'-a_3}, \end{aligned}$$

via the isomorphism

$$\begin{aligned} X'=X,\quad Y'=a_1X+Y+a_3Z,\quad Z'=Z. \end{aligned}$$

Note that the new coordinates \([X':Y':Z']\) are the coordinates of the negative of the point [X : Y : Z].

The new coordinate on the formal group is

$$\begin{aligned} x'=g(x)=\frac{x}{1+a_1x+a_3z(x)}. \end{aligned}$$

But since we have just taken the coordinate of the negative point on the elliptic curve, we also have

$$\begin{aligned} g(x)=\overline{x}, \end{aligned}$$

since the inverse on the formal group corresponds to the Chern class of the complex conjugate of the canonical bundle over BU(1), i.e. \(\overline{x}=[-1](x)\). Note that this implies \(g(g(x))=x\).

We can consider both coordinates x and \(x'=\overline{x}\) as ring maps \(MU\rightarrow { TMF }_1(3)\). By composition with \(MU\langle 6 \rangle \rightarrow MU\), we obtain \(x,x': MU\langle 6 \rangle \rightarrow { TMF }_1(3)\), and together with \(\sigma :MU\langle 6 \rangle \rightarrow { TMF }_1(3)\), we obtain \(r_U=\frac{\sigma }{x},r_U'=\frac{\sigma }{x'}:BU\langle 6 \rangle _+\rightarrow { TMF }_1(3)\). Let \(\tau \) denote the involutions on \(MU\langle 6 \rangle ,BU\langle 6 \rangle _+, { TMF }_1(3)\) respectively. From our considerations, it follows that the left diagram

commutes up to homotopy: the three maps \(MU\rightarrow { TMF }_1(3)\) correspond to the element \(x'=g(x)=\overline{x}\). The fact that the right diagram commutes up to homotopy follows from the construction of \(\sigma \) in [3]. Therefore \(r_U:BU\langle 6 \rangle _+\rightarrow { TMF }_1(3)\) is up to homotopy invariant—both compositions \(\tau \circ r_U\) and \(r_U\circ \tau \) are homotopic to \(r'_U\).

In the remaining part of this appendix, we consider maximal tori. We show that the Pontryagin classes are invariant, and we also show (again) that \(r_U\) has the same image in \({ TMF }_1(3)^*{ BString}\) as \(r'_U=r_U\cdot \frac{x}{x'}\). (Both x and \(x'\) are generators for the free rank one \({ TMF }_1(3)^*BU\)-module \({ TMF }_1(3)^*{ MU}\), so that there is a unique invertible element \(\frac{x}{x'}\) of \({ TMF }_1(3)^*{ BU}\) whose product with \(x'\) is x.)

Set \(E={ TMF }_1(3)\). Using the complex coordinate x, we have isomorphisms and an injection

$$\begin{aligned} E^*({ BU})\cong E^*\llbracket c_1,c_2,\ldots \rrbracket \rightarrow E^*({ BU}(1)^\infty )\cong E^*\llbracket y_1,y_2,\ldots \rrbracket , \end{aligned}$$

where \(BU(1)^\infty ={{\mathrm{\mathrm{colim}}}}_N BU(1)^N\) and where each \(c_k\) is mapped to the kth elementary symmetric polynomial in the \(y_i\). This is induced by the map

$$\begin{aligned} \sum (L_i-1): BU(1)^\infty \rightarrow BU. \end{aligned}$$

Let

$$\begin{aligned} Q(x)= \frac{x}{g(x)}=1+a_1x+a_3z(x)\in E^*\llbracket x\rrbracket . \end{aligned}$$

By Hirzebruch’s theory of multiplicative sequences, in \(E^*\llbracket y_1,y_2,\ldots \rrbracket \) the element \(\frac{x}{x'}\) corresponds to \(\prod _{k=1}^{\infty }Q(y_k)\), which is symmetric in the \(y_k\), so that it defines an element in \(E^*\llbracket \sigma _j(y_1,y_2,\ldots ) \mid j\ge 1\rrbracket \cong E^*{ BU}\). Here \(\sigma _j(y_1,y_2,\ldots )\) is the jth elementary symmetric polynomial of the \(y_i\).

We now consider the restriction to \( BString \) under the map \(BString\rightarrow BSO\mathop {\rightarrow }\limits ^{c}BU\). Unstably, we consider a maximal torus \({\mathbb T}'\cong U(1)^{2N+1}\) in \(U(2N+1)\) and its Weyl group \(W\cong \Sigma _{2N+1}\). We have

$$\begin{aligned} E^*B{\mathbb T}'\cong E^*\llbracket y_1,\ldots , y_{2N+1}\rrbracket , \end{aligned}$$

where \(y_i\) is the Chern class of the canonical line bundle \(L_i\) over the ith factor, and

$$\begin{aligned} E^*{ BU}(2N+1)\cong E^*\llbracket c_1,c_2,\ldots ,c_{2N+1}\rrbracket \cong E^*B{\mathbb T}'^W, \end{aligned}$$

where W acts on \(E^*\llbracket y_1,\ldots ,y_{2N+1}\rrbracket \) by permuting the \(y_i\). The Chern classes \(c_j=\sigma _j(y_1,y_2,\ldots )\) are the elementary symmetric polynomials of the \(y_i\).

Comparing the unitary and special orthogonal groups and their maximal tori, we obtain the diagram

The maximal torus in \(SO(2N+1)\) consists of matrices

$$\begin{aligned} A=\begin{pmatrix} R_{\phi _1}&{}\quad &{}\quad &{}\\ {} &{}\quad \ldots &{}\quad &{}\\ {} &{}\quad &{}R_{\phi _N}&{}\\ {} &{}\quad &{}\quad &{}1\end{pmatrix}, \end{aligned}$$

where \(R_\phi =\begin{pmatrix}\cos \phi &{}-\sin \phi \\ \sin \phi &{}\cos \phi \end{pmatrix}\). The standard maximal torus in U(N) consists of diagonal matrices. Under conjugation by

$$\begin{aligned} \begin{pmatrix} T&{}\quad &{}\quad &{}\\ {} &{}\ldots &{}\quad &{}\\ {} &{}\quad &{}T&{}\\ {} &{}\quad &{}\quad &{}1\end{pmatrix}, \end{aligned}$$

where \(T=\frac{1}{\sqrt{2}}\begin{pmatrix}1&1{\mathbf i}&i\end{pmatrix},\) the matrix A corresponds to the diagonal matrix \(diag(e^{i\phi _1},e^{-i\phi _1},e^{i\phi _2},e^{-i\phi _2},\ldots ,1)\), so that we can assume that the map \(B{\mathbb T}\rightarrow B{\mathbb T}'\) is induced by

$$\begin{aligned} U(1)^N\rightarrow U(1)^{2N+1},\quad (z_1,\ldots ,z_N)\mapsto (z_1,z_1^{-1}, \ldots , z_N,z_N^{-1},1). \end{aligned}$$

On the classifying spaces, the induced map \(BU(1)^N\rightarrow BU(1)^{2N+1}\) pulls back \(L_{2k-1}\) to \(L_k\) and \(L_{2k}\) to \(\overline{L_k}\) for \(k\le N\), while the pullback of \(L_{2N+1}\) is trivial. It follows that in cohomology,

$$\begin{aligned} y_{2k-1}\mapsto x_k,\qquad y_{2k}\mapsto \overline{x_k}=g(x_k)\ \ \ \text {for } k\le N\ \ \text { and} \quad y_{2N+1}\mapsto 0. \end{aligned}$$

For the image of the class \(\frac{x}{x'}\), we have

$$\begin{aligned} Q(y_{2k-1})\mapsto Q(x_k)=\frac{x_k}{g(x_k)},\quad Q(y_{2k})\mapsto Q(g(x_k))=\frac{g(x_k)}{g(g(x_k))}=\frac{g(x_k)}{x_k} \end{aligned}$$

and \(Q(y_{2N+1})\mapsto 1\). It follows that \(\prod _{k=1}^{\infty }Q(y_k)\mapsto 1\).

We conclude that the image of \(\frac{x}{x'}\) in \(E^*{ BString}\) is trivial, so that \({\mathbb Z}/2\) acts trivially on \(r\in E^*({ BString})\).

Since the action of \({\mathbb Z}/2\) on \(E^*B{\mathbb T}\) sends each \(x_i\mapsto \overline{x_i}\), we see that all \(x_i\cdot \overline{x_i}\) and therefore also all Pontryagin classes are invariant elements of \(E^*({ BString})\).

Appendix 2: Computation of a coefficient of the cubical structure

We prove the part of Proposition 3.3 which is used in the rest of the paper. The calculation uses the notation introduced in Sect. 3.

Proposition 6.1

Modulo the ideal \((2,a_1,a_2)\), the cubical structure corresponding to \(r_U\) has the form \(1+a_3x_0x_1x_2+\text {terms of higher total degree}\).

Proof

We compute modulo terms of higher order and use the notation \(\sum \nolimits _{ cyc }\) for summation over cyclic permutations of the indices 0, 1, 2 and \(\sum \nolimits _{{ sym}}\) for summation over all permutations:

$$\begin{aligned} r_U&=\frac{t}{u} = \frac{\left|\begin{matrix}x_0&{}z_0\\ x_1 &{} z_1\end{matrix}\right|\left|\begin{matrix}x_1&{}z_1\\ x_2 &{} z_2\end{matrix}\right|\left|\begin{matrix}x_2&{}z_2\\ x_0 &{} z_0\end{matrix}\right|}{\left|\begin{matrix}x_0&{}1&{}z_0\\ x_1 &{} 1 &{} z_1\\ x_2 &{} 1 &{} z_2\end{matrix}\right|z_0z_1z_2} \cdot \frac{x_0x_1x_2(x_0+_F x_1+_Fx_2)}{(x_0+_F x_1)(x_1+_F x_2)(x_2+_F x_0)}\\ \\&=\frac{\prod _{ cyc }(x_0\cdot z(x_1)-x_1\cdot z(x_0))\cdot \prod _{ cyc }x_0 \cdot (x_0+_F x_1+_F x_2) }{\left( \sum _{ cyc }x_1\cdot z(x_0)-x_0\cdot z(x_1) \right) \cdot \prod _{ cyc }z(x_0)\cdot \prod _{ cyc }(x_0+_Fx_1)}\\&=\frac{\prod _{ cyc }\left( x_1^2+a_3x_1^5-x_0^2-a_3x_0^5\right) \cdot \left( \sum _{ cyc }x_0+\sum _{ cyc } a_3x_0^2x_1^2\right) }{\left( \sum _{ cyc }x_1x_0^3+a_3x_1x_0^6-x_0x_1^3-a_3x_0x_1^6\right) \cdot \prod _{ cyc }(1+a_3x_0^3)\cdot \prod _{ cyc }\left( x_0+x_1+a_3x_0^2x_1^2\right) }, \end{aligned}$$

where we have divided numerator and denominator by \(x_0^3x_1^3x_2^3\) in the last step. In the resulting fraction, the two terms of lowest order of the numerator have the form

$$\begin{aligned}&\prod _{ cyc }\left( x_1^2-x_0^2\right) \cdot \left( \sum _{ cyc }x_0\right) + \prod _{ cyc }\left( x_1^2-x_0^2\right) \cdot \left( \sum _{ cyc } a_3x_0^2x_1^2\right) \\&\quad +\,\left( \sum _{ cyc } a_3\left( x_1^5-x_0^5\right) \left( x_2^2-x_1^2\right) \left( x_0^2-x_2^2\right) \right) \cdot \left( \sum _{ cyc }x_0\right) , \end{aligned}$$

and for the denominator we obtain the two lowest order terms

$$\begin{aligned}&\left( \sum _{ cyc }x_1x_0^3-x_0x_1^3 \right) \cdot \prod _{ cyc }(x_0+x_1)\\&\qquad +\left( \sum _{ cyc }x_1x_0^3-x_0x_1^3 \right) \cdot \left( \sum _{ cyc }(x_0+x_1)(x_1+x_2)a_3x_0^2x_2^2\right) \!\\&\qquad +\! \left( \sum _{ cyc }x_1x_0^3\!-\!x_0x_1^3 \right) \cdot \left( \sum _{ cyc }a_3x_0^3\right) \prod _{ cyc }(x_0+x_1)\\&\qquad +\left( \sum _{ cyc }a_3(x_1x_0^6-x_0x_1^6) \right) \cdot \prod _{ cyc }(x_0+x_1). \end{aligned}$$

Note that the leading terms of numerator and denominator agree, so that the quotient has the form \(1+\frac{v}{w}\) where v is the difference of the terms of second lowest order in numerator and denominator and w is the common term of lowest order. Now we do the computation modulo 2 and obtain the equality

$$\begin{aligned} \frac{v}{a_3}= & {} \left( \sum _{{ sym}}x_0^4x_1^2 \right) \cdot \left( \sum _{ cyc } x_0^2x_1^2\right) + \left( \sum _{{ sym}}x_0^5x_1^4\right) \cdot \left( \sum _{ cyc } x_0\right) \\&+\left( \sum _{{ sym}}x_0^3x_1\right) \cdot \left( x_0^2x_1^2x_2^2 + \sum _{ cyc } x_0^3x_1^3+ \sum _{{ sym}}x_0^3x_1^2x_2\right) \\&+ \left( \sum _{{ sym}}x_0^3x_1\right) \cdot \left( \sum _{{ sym}}x_0^5x_1+x_0^4x_1^2+x_0^3x_1^2x_2\right) \\&+\left( \sum _{{ sym}} x_0^8x_1^2+x_0^7x_1^3+x_0^7x_1^2x_2+x_0^6x_1^3x_2\right) \\= & {} \sum _{{ sym}} x_0^5x_1^4x_2+ x_0^5x_1^3x_2^2+ x_0^6x_1^3x_2 = x_0x_1x_2 \cdot \left( \sum _{{ sym}} x_0^4x_1^3+ x_0^4x_1^2x_2^1+ x_0^5x_1^2\right) \\= & {} x_0x_1x_2 \cdot \left( \sum _{{ sym}} x_0^4x_1^2\right) \cdot \left( \sum _{ cyc }x_0\right) =x_0x_1x_2w. \end{aligned}$$

\(\square \)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Laures, G., Olbermann, M. \({ TMF }_0(3)\)-Characteristic classes for string bundles. Math. Z. 282, 511–533 (2016). https://doi.org/10.1007/s00209-015-1551-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00209-015-1551-3

Mathematics Subject Classification

Navigation