Algebraic Topology Mid-Semestral Examination

Algebraic Topology Mid-Semestral Examination

Algebraic Topology M.Math II Mid-Semestral Examination http://www.isibang.ac.in/∼statmath/oldqp/WWWW Midterm Solution (1). Decide whether the following statements are TRUE or FALSE. Justify your answers. State precisely the results that you use. (a) The pull back of a trivial bundle is trivial. (b) The tangent bundle of S2 splits as the Whitney sum of two line bundles. (c) !(γ1) = 1 + a. (d) There exists a vector bundle η such that γ1 ⊕ η is trivial. 2 3 (e) The manifolds RP × RP and S5 are cobordant. Solution: (a) True Justification: Let ' : E ! B × V be a trivialization of a vector bundle p : E ! B. Let f : X ! B be Id×' a continuous map. Then : f ∗E ⊆ X × E −! X × B × V ! X × V is a trivialization of f ∗E with f×Id '−1 inverse induced by the projection g1 : X × V ! X and the map g2 : X × V −! B × V −! E. (b)False Justification: Theorem 1 (Classification of line bundles) The association E ! !1(E) gives a bijection between iso- 1 morphism classes of real line bundles on B and H (B; Z2). Theorem 2 (Hairy Ball Theorem) There is no nonvanishing continuous tangent vector field on even dimensional n-spheres. 2 2 2 Suppose the tangent bundle of S , T S = L1 ⊕ L2, where L1 and L2 are line bundles over S . By the classification of line bundles, L1 and L2 are trivial bundles. This implies that there is a nonvanishing continuous tangent vector field on S2. This is a contradiction by Hairy Ball Theorem, and therefore our assumption is wrong. (c) True 1 Justification: Let γ1 be the canonical line bundle over RP . Then we have an obvious bundle map 1 1 E(γ1 ) −−−−! E(γ ) ? ? ?γ1 ? 1 (1) y 1 yγ 1 inclus 1 S1 = RP −−−−! RP ∗ 1 ∗ 1 1 Therefore i (!1(γ )) = !1(i (γ )) = !1(γ1 ) 6= 0, by the axiom of the Stiefel-Whitney classes. This shows 1 1 1 ∼ 1 1 1 that !1(γ ) = a is the generator of H (RP ; Z2) = Z2. Hence !(γ ) = !0(γ ) + !1(γ ) = 1 + a, since the bundle is one dimensional. (d) False 1 Justification: Suppose there exists a bundle η : E ! RP such that γ1 ⊕η is trivial. Since !(γ1) = 1+a, 1 1 1 −1 2 where a is the generator of H (RP ; Z2) and hence !(E) must be (1 + a) = 1 + a + a + ··· . Since i i 1 we are using Z2 coefficients. Thus !i(E) = a , which is non-zero in H (RP ; Z2)for all i. However, this contradicts the fact that !i(E) = 0 for i > dim E. (e) False Justification: Theorem 3 (Pontrjagin) If B is a smooth compact (n + 1)-dimensional manifold with boundary equal to M, then the Stiefel-Whitney numbers of M are all zero. Theorem 4 Two smooth closed n-manifolds belong to the same cobordism class if and only if all of their corresponding Stiefel-Whitney numbers are equal. 2 3 5 2 3 2 3 Claim: RP × RP and S are not cobordant. Since !1(RP × RP ) = !1(RP ) [ !0(RP ) = 3a [ 1 6= 0 1 2 3 1 2 2 3 in H (RP × RP ; Z2), where a is the generator of H (RP ; Z2) of RP × RP . So not all Stiefel-Whitney 2 3 numbers are 0. But S5 = @D6. So, by Theorem 3 and Theorem 4, RP × RP and S5 are not cobordant. (2). 9 (a). Determine the least integer k assuming that RP can be immersed in R9+k. Solution: Now suppose that a manifold M n of dimension n can be immersed in Rn+k. Recall that a n n+k n n+k function f : M ! R is an immersion if at every point x 2 M , Dfx : TxM ! R is injective. This means that the tangent bundle TM is an n-dimensional vector bundle, and the normal bundle ν n is a k-dimensional vector bundle. From the Whitney Duality theorem, !i(ν) = !i(M ). So the dual n 9 Stiefel-Whitney classes !i(M ) = 0 for i > k. For M = RP , 9 2 8 !(RP ) = 1 + a + a and 9 2 4 6 !(RP ) = 1 + a + a + a : 9 So if RP can be immersed in R9+k, then k must be atleast 6. n (b). Show that the tangent bundle of RP has a sub bundle of rank 1 if and only if n is odd. n n Solution: Since !(RP ) = (1 + a)n+1. If T (RP ) = ξ ⊕ η for a line bundle ξ, then we must have n !(ξ)!(η) = !(RP ) = (1 + a)n+1. !(ξ) and !(η) must be order 1 and n − 1 respectively(by the axioms of Stiefel-Whitney classes) so it must be the case that !(ξ) = 1 + a or 1. If n is even, then in the first case, n n n n n+1 n n !(η) = (1 + a) has !n(η) = n a = a , and in the later case !n(η) = n a = a . This contradicts the fact that for an (n − 1)-bundle such as η, !n(η) = 0. Thus such a splitting cannot happen. If n is n n odd, then !n(RP ) = 0. This implies that T (RP ) admits one global section by whose span defines a sub-bundle of dimension 1. n (c). Let n = 2k. Find all the Stiefel-Whitney numbers of RP . n Solution: !(RP ) = (1 + a)n+1 = (1 + a)(1 + a)n. Since we are working modulo 2 and n = 2k. We have n n (1+a)n = 1+an. So !(RP ) = (1+a)(1+a)n = 1+a+an +an+1 = 1+a+an. Thus !(RP ) = 1+a+an. n n n n n This implies !0(RP ) = 1, !1(RP ) = a, !n(RP ) = a and !i(RP ) = 0 8 i 6= 0; 1; n. So the Stiefel- n n n n n n n ∼ Whitney numbers h!n(RP ); [RP ]i and h!1(RP ) ; [RP ]i are the only non-zero in H (RP ; Z2) = Z2. All other Stiefel-Whitney numbers are zero. 2 1 1 k (d). Let k ≥ 1 be an integer and fk : S ! S be the power map z 7! z . For which values of k is the ∗ 1 bundle fk (γ1 ) trivial? 1 1 1 1 1 Solution: Let γ1 be the canonical line bundle over RP = S . Since fk : S ! S has degree k, then ∗ 1 1 1 1 ∗ ∗ 1 1 fk : H (S ; Z2) ! H (S ; Z2) such that fk (x) = kx. Suppose the pull back bundle p : fk (γ1 ) ! S is ∗ 1 ∗ 1 ∗ 1 trivial.This implies that 0 = !1(fk (γ1 )) = fk (!(γ1 )) = fk (a) = ka where a = !1(γ1 ) is the generator of 1 1 ∼ H (S ; Z2) = Z2. This shows that k ≡ 0 mod 2 3.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    3 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us