Exotic sphere

In differential topology, a mathematical discipline, an exotic sphere is a differentiable manifold M that is homeomorphic but not diffeomorphic to the standard Euclidean n-sphere. That is, M is a sphere from the point of view of all its topological properties, but carrying a smooth structure that is not the familiar one (hence the name "exotic").

The first exotic spheres were constructed by John Milnor (1956) in dimension n = 7 as S³-bundles over S⁴. He showed that there are at least 7 differentiable structures on the 7-sphere. In any dimension Milnor (1959) showed that the diffeomorphism classes of oriented exotic spheres form the non-trivial elements of an abelian monoid under connected sum, which is a finite abelian group if the dimension is not 4. The classification of exotic spheres by Michel Kervaire and John Milnor (1963) showed that the oriented exotic 7-spheres are the non-trivial elements of a cyclic group of order 28 under the operation of connected sum.

Introduction

The unit n-sphere, Sⁿ, is the set of all n+1-tuples (x₁, x₂, ... x_n+1) of real numbers, such that the sum x₁² + x₂² + ... + x_n+1² = 1. (S¹ is a circle; S² is the surface of an ordinary ball of radius one in 3 dimensions.) Topologists consider a space, X, to be an n-sphere if every point in X can be assigned to exactly one point in the unit n-sphere in a continuous way, which means that sufficiently nearby points in X get assigned to nearby points in Sⁿ and vice-versa. For example a point x on an n-sphere of radius r can be matched with a point on the unit n-sphere by adjusting its distance from the origin by 1/r.

In differential topology, a more stringent condition is added, that the functions matching points in X with points in Sⁿ should be smooth, that is they should have derivatives of all orders everywhere. To calculate derivatives, one needs to have local coordinate systems defined consistently in X. Mathematicians were surprised in 1956 when John Milnor showed that consistent coordinate systems could be set up on the 7-sphere in two different ways that were equivalent in the continuous sense, but not in the differentiable sense. Milnor and others set about trying to discover how many such exotic spheres could exist in each dimension and to understand how they relate to each other. No exotic structures are possible on the 1-, 2-, 3-, 5-, 6- or 12-spheres. Some higher dimensional spheres have only two possible differentiable structures, others have thousands. Whether exotic 4-spheres exist, and if so how many, is an important unsolved problem in mathematics.

The monoid of smooth structures on spheres in a given dimension

The monoid of smooth structures on n-spheres is the collection of oriented smooth n-manifolds which are homeomorphic to the n-sphere, taken up to orientation-preserving diffeomorphism. The monoid operation is the connected sum. Provided n ≠ 4, this monoid is a group and is isomorphic to the group Θ_n of h-cobordism classes of oriented homotopy n-spheres, which is finite and abelian. In dimension 4 almost nothing is known about the monoid of smooth spheres, beyond the facts that it is finite or countably infinite, and abelian, though it is suspected to be infinite; see the section on Gluck twists. All homotopy n-spheres are homeomorphic to the n-sphere by the generalized Poincaré conjecture, proved by Stephen Smale in dimensions bigger than 4, Michael Freedman in dimension 4, and Grigori Perelman in dimension 3. In dimension 3, Edwin E. Moise proved that every topological manifold has an essentially unique smooth structure (see Moise's theorem), so the monoid of smooth structures on the 3-sphere is trivial.

The group Θ_n has a cyclic subgroup

$bP_{n+1}\$

represented by n-spheres that bound parallelizable manifolds. The structures of bP_n+1 and the quotient

$\Theta_n/bP_{n+1}\$

are described separately in the paper (Michel Kervaire & John Milnor 1963), which was influential in the development of surgery theory. In fact, these calculations can be formulated in a modern language in terms of the surgery exact sequence as indicated here.

The group bP_n+1 is trivial if n is even. If n is 1 mod 4 it has order 1 or 2; in particular it has order 1 if n is 1, 5, 13, 29, or 61, and Browder (1969) proved that it has order 2 if n = 1 mod 4 is not of the form 2^k – 3. The order of bP_4n for n ≥ 2 is

$2^{2n-2}(2^{2n-1}-1)B \,\!$

where B is the numerator of |4B_2n/n|, and B_2n is a Bernoulli number. (The formula in the topological literature differs slightly because topologists use a different convention for naming Bernoulli numbers; this article uses the number theorists' convention.)

The quotient group Θ_n/bP_n+1 has a description in terms of stable homotopy groups of spheres modulo the image of the J-homomorphism). More precisely there is an injective map

$\Theta_n/bP_{n+1}\to \pi_n^S/J\,$

where π_n^S is the nth stable homotopy group of spheres, and J is the image of the J-homomorphism. Browder (1969) proved this is an isomorphism if n is not of the form 2^k – 2, and if n is of this form its image is either the whole group or a subgroup of index 2, and is a subgroup of index 2 in the first few cases when n is 2, 6, 14, 30, or 62. This index is 2 if and only if there exists an n-dimensional framed manifold with Kervaire invariant 1. Finding the dimensions n in which this index is 2 is the Kervaire invariant problem, whose solution for all larger n except 126 was announced by Mike Hill, Michael Hopkins and Doug Ravenel: the above list of dimensions can be extended by at most one, at most including 126.

The order of the group Θ_n is given in this table (sequence A001676 in OEIS) from (Kervaire & Milnor 1963) (except that the entry for n = 19 is wrong by a factor of 2 in their paper; see the correction in volume III p. 97 of Milnor's collected works).

Dim n	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20
order Θn	1	1	1	1	1	1	28	2	8	6	992	1	3	2	16256	2	16	16	523264	24
bPn+1	1	1	1	1	1	1	28	1	2	1	992	1	1	1	8128	1	2	1	261632	1
Θn/bPn+1	1	1	1	1	1	1	1	2	2×2	6	1	1	3	2	2	2	2×2×2	8×2	2	24
πnS/J	1	2	1	1	1	2	1	2	2×2	6	1	1	3	2×2	2	2	2×2×2	8×2	2	24

Further entries in this table can be computed from the information above together with the table of stable homotopy groups of spheres.

Explicit examples of exotic spheres

When I came upon such an example in the mid-50s, I was very puzzled and didn’t know what to make of it. At first, I thought I’d found a counterexample to the generalized Poincaré conjecture in dimension seven. But careful study showed that the manifold really was homeomorphic to S⁷. Thus, there exists a differentiable structure on S⁷ not diffeomorphic to the standard one.

John Milnor (2009, p.12)

One of the first examples of an exotic sphere found by Milnor (1956, section 3) was the following: Take two copies of B⁴×S³, each with boundary S³×S³, and glue them together by identifying (a,b) in the boundary with (a, a²ba⁻¹), (where we identify each S³ with the group of unit quaternions). The resulting manifold has a natural smooth structure and is homeomorphic to S⁷, but is not diffeomorphic to S⁷. Milnor showed that it is not the boundary of any smooth 8-manifold with vanishing 4th Betti number, and has no orientation-reversing diffeomorphism to itself; either of these properties implies that it is not a standard 7-sphere. Milnor showed that this manifold has a Morse function with just two critical points, both non-degenerate, which implies that it is topologically a sphere.

As shown by Egbert Brieskorn (1966, 1966b) (see also (Hirzebruch & Mayer 1968)) the intersection of the complex manifold of points in C⁵ satisfying

$a^2 + b^2 + c^2 + d^3 + e^{6k-1} = 0\$

with a small sphere around the origin for k = 1, 2, ..., 28 gives all 28 possible smooth structures on the oriented 7-sphere.

Twisted spheres

Given an (orientation-preserving) diffeomorphism f: Sⁿ⁻¹→Sⁿ⁻¹, gluing the boundaries of two copies of the standard disk Dⁿ together by f yields a manifold called a twisted sphere (with twist f). It is homotopy equivalent to the standard n-sphere because the gluing map is homotopic to the identity (being an orientation-preserving diffeomorphism, hence degree 1), but not in general diffeomorphic to the standard sphere. (Milnor 1959b) Setting Γ_n to be the group of twisted n-spheres (under connect sum), one obtains the exact sequence

$\pi_0\,\text{Diff}^+(D^n) \to \pi_0\,\text{Diff}^+(S^{n-1}) \to \Gamma_n \to 0. \,\!$

For n > 4, every exotic sphere is diffeomorphic to a twisted sphere, a result proven by Stephen Smale. (In contrast, in the piecewise linear setting the left-most map is onto via radial extension: there are no piecewise-linear-twisted spheres.) The group Γ_n of twisted spheres is always isomorphic to the group Θ_n. The notations are different because it was not known at first that they were the same for n=3 or 4; for example, the case n=3 is equivalent to the Poincaré conjecture.

In 1970 Jean Cerf proved the pseudoisotopy theorem which implies that $\pi_0\,\text{Diff}^+(D^n)$ is the trivial group provided $n \geq 6$, so $\Gamma_n \simeq \pi_0\,\text{Diff}^+(S^{n-1})$ provided $n \geq 6$.

Applications

If M is a piecewise linear manifold then the problem of finding the compatible smooth structures on M depends on knowledge of the groups Γ_k = Θ_k. More precisely, the obstructions to the existence of any smooth structure lie in the groups H_k+1(M, Γ_k) for various values of k, while if such a smooth structure exists then all such smooth structures can be classified using the groups H_k(M, Γ_k). In particular the groups Γ_k vanish if k<7, so all PL manifolds of dimension at most 7 have a smooth structure, which is essentially unique if the manifold has dimension at most 6.

The following finite abelian groups are essentially the same:

• The group Θ_n of h-cobordism classes of oriented homotopy n-spheres.
• The group of h-cobordism classes of oriented n-spheres.
• The group Γ_n of twisted oriented n spheres.
• The homotopy group π_n(PL/DIFF)
• If n ≠ 3, the homotopy π_n(TOP/DIFF) (if n=3 this group has order 2; see Kirby–Siebenmann invariant).
• The group of smooth structures of an oriented PL n-sphere.
• If n≠4, the group of smooth structures of an oriented topological n-sphere.
• If n≠5, the group of components of the group of all orientation-preserving diffeomorphisms of Sⁿ⁻¹.

4-dimensional exotic spheres and Gluck twists

In 4 dimensions it is not known whether there are any exotic smooth structures on the 4-sphere. The statement that they do not exist is known as the "smooth Poincaré conjecture", and is discussed by Michael Freedman, Robert Gompf, and Scott Morrison et al. (2010) who say that it is believed to be false.

Some candidates for exotic 4-spheres are given by Gluck twists (Gluck 1962). These are constructed by cutting out a tubular neighborhood of a 2-sphere S in S⁴ and gluing it back in using a diffeomorphism of its boundary S²×S¹. The result is always homeomorphic to S⁴. But in most cases it is unknown whether or not the result is diffeomorphic to S⁴. (If the 2-sphere is unknotted, or given by spinning a knot in the 3-sphere, then the Gluck twist is known to be diffeomorphic to S⁴, but there are plenty of other ways to knot a 2-sphere in S⁴.)

Akbulut (2009) showed that a certain family of candidates for 4-dimensional exotic spheres constructed by Cappell and Shaneson are in fact standard.

See also

• Atlas (topology)
• Clutching construction
• Exotic R⁴
• Cerf theory
• Seven-dimensional space

References

• Akbulut, Selman (2009), Cappell–Shaneson homotopy spheres are standard, arXiv:0907.0136 
• Brieskorn, Egbert V. (1966), "Examples of singular normal complex spaces which are topological manifolds", Proceedings of the National Academy of Sciences 55 (6): 1395–1397, doi:10.1073/pnas.55.6.1395, MR0198497 
• Brieskorn, Egbert (1966b), "Beispiele zur Differentialtopologie von Singularitäten", Invent. Math. 2 (1): 1–14, doi:10.1007/BF01403388, MR0206972 
• Browder, William (1969), "The Kervaire invariant of framed manifolds and its generalization", Annals of Mathematics 90 (1): 157–186, doi:10.2307/1970686, JSTOR 1970686, MR0251736 
• Freedman, Michael; Gompf, Robert; Morrison, Scott; Walker, Kevin (2010), "Man and machine thinking about the smooth 4-dimensional Poincaré conjecture", Quantum Topology 1 (2): 171–208, arXiv:0906.5177 
• Gluck, Herman (1962), "The embedding of two-spheres in the four-sphere", Transactions of the American Mathematical Society 104 (2): 308–333, doi:10.2307/1993581, JSTOR 1993581, MR0146807 
• Hirzebruch, Friedrich; Mayer, Karl Heinz (1968), O(n)-Mannigfaligkeiten, Exotische Sphären und Singularitäten, Lecture Notes in Mathematics, 57, Berlin-New York: Springer-Verlag, doi:10.1007/BFb0074355, MR0229251  This book describes Brieskorn's work relating exotic spheres to singularities of complex manifolds.
• Kervaire, Michel A.; Milnor, John W. (1963), "Groups of homotopy spheres: I", Annals of Mathematics 77 (3): 504–537, doi:10.2307/1970128, http://www.uni-math.gwdg.de/schick/publ/Groups%20of%20homotopy%20spheres%20I.pdf  This paper describes the structure of the group of smooth structures on an n-sphere for n > 4. Sadly, the promised paper "Groups of Homotopy Spheres: II" never appeared, but Levine's lecture notes contain the material which it might have been expected to contain.
• Levine, J.P. (1985), "Lectures on groups of homotopy spheres", Algebraic and geometric topology, Lecture Notes in Mathematics, 1126, Berlin-New York: Springer-Verlag, pp. 62–95, doi:10.1007/BFb0074439, MR8757031 
• Milnor, John W. (1956), "On manifolds homeomorphic to the 7-sphere", Annals of Mathematics 64 (2): 399–405, doi:10.2307/1969983, JSTOR 1969983, MR0082103 
• Milnor, John W. (1959), "Sommes de variétes différentiables et structures différentiables des sphères", Bulletin de la Société Mathématique de France 87: 439–444, MR0117744, http://www.numdam.org/item?id=BSMF_1959__87__439_0 
• Milnor, John W. (1959b), "Differentiable structures on spheres", American Journal of Mathematics 81 (4): 962–972, doi:10.2307/2372998, JSTOR 2372998, MR0110107 
• Milnor, John (2000), Classification of (n − 1)-connected 2n-dimensional manifolds and the discovery of exotic spheres., "Surveys on surgery theory", Annals of Mathematics Studies 145: 25–30, MR1747528 .
• Milnor, John Willard (2009), "Fifty years ago: topology of manifolds in the 50's and 60's", in Mrowka, Tomasz S.; Ozsváth., Peter S., Low dimensional topology. Lecture notes from the 15th Park City Mathematics Institute (PCMI) Graduate Summer School held in Park City, UT, Summer 2006., IAS/Park City Math. Ser., 15, Providence, R.I.: Amer. Math. Soc., pp. 9–20, ISBN 978-0-8218-4766-4, MR2503491, http://www.math.sunysb.edu/~jack/PREPRINTS/pcity-lec.pdf 
• Milnor, John W. (2011), "Differential topology forty-six years later", Notices of the American Mathematical Society 58 (6): 804–809, http://www.ams.org/notices/201106/rtx110600804p.pdf 
• Rudyak, Yu.B. (2001), "Milnor sphere", in Hazewinkel, Michiel, Encyclopaedia of Mathematics, Springer, ISBN 978-1556080104, http://eom.springer.de/M/m063800.htm 

External links

• Exotic spheres on the Manifold Atlas.
• Exotic sphere home page on the home page of Andrew Ranicki. Assorted source material relating to exotic spheres.