Exotic sphere

The unit n-sphere, ${\displaystyle S^{n}}$ , is the set of all (n+1)-tuples ${\displaystyle (x_{1},x_{2},\ldots ,x_{n+1})}$  of real numbers, such that the sum ${\displaystyle x_{1}^{2}+x_{2}^{2}+\cdots +x_{n+1}^{2}=1}$ . For instance, ${\displaystyle S^{1}}$  is a circle, while ${\displaystyle S^{2}}$  is the surface of an ordinary ball of radius one in 3 dimensions. Topologists consider a space X to be an n-sphere if there is a homeomorphism between them, i.e. every point in X may be assigned to exactly one point in the unit n-sphere by a continuous bijection with continuous inverse. For example, a point x on an n-sphere of radius r can be matched homeomorphically with a point on the unit n-sphere by multiplying its distance from the origin by ${\displaystyle 1/r}$ . Similarly, an n-cube of any radius is homeomorphic to an n-sphere.

In differential topology, two smooth manifolds are considered smoothly equivalent if there exists a diffeomorphism from one to the other, which is a homeomorphism between them, with the additional condition that it be smooth — that is, it should have derivatives of all orders at all its points — and its inverse homeomorphism must also be smooth. To calculate derivatives, one needs to have local coordinate systems defined consistently in X. Mathematicians (including Milnor himself) were surprised in 1956 when Milnor showed that consistent local 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-, 12-, 56- or 61-sphere.^[1] 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 unsolved problem.

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 ${\displaystyle n\neq 4}$ , this monoid is a group and is isomorphic to the group ${\displaystyle \Theta _{n}}$  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 ${\displaystyle \Theta _{n}}$  has a cyclic subgroup

${\displaystyle bP_{n+1}}$ 

represented by n-spheres that bound parallelizable manifolds. The structures of ${\displaystyle bP_{n+1}}$  and the quotient

${\displaystyle \Theta _{n}/bP_{n+1}}$ 

are described separately in the paper (Kervaire & 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 ${\displaystyle bP_{n+1}}$  is a cyclic group, and is trivial or order 2 except in case ${\displaystyle n=4k+3}$ , in which case it can be large, with its order related to the Bernoulli numbers. It 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 William Browder (1969) proved that it has order 2 if ${\displaystyle n=1}$  mod 4 is not of the form ${\displaystyle 2^{k}-3}$ . It follows from the now almost completely resolved Kervaire invariant problem that it has order 2 for all n bigger than 126; the case ${\displaystyle n=126}$  is still open. The order of ${\displaystyle bP_{4k}}$  for ${\displaystyle k\geq 2}$  is

${\displaystyle 2^{2k-2}(2^{2k-1}-1)B,}$ 

where B is the numerator of ${\displaystyle 4B_{2k}/k}$ , and ${\displaystyle B_{2k}}$  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 ${\displaystyle \Theta _{n}/bP_{n+1}}$  has a description in terms of stable homotopy groups of spheres modulo the image of the J-homomorphism; it is either equal to the quotient or index 2. More precisely there is an injective map

${\displaystyle \Theta _{n}/bP_{n+1}\to \pi _{n}^{S}/J,}$ 

where ${\displaystyle \pi _{n}^{S}}$  is the nth stable homotopy group of spheres, and J is the image of the J-homomorphism. As with ${\displaystyle bP_{n+1}}$ , the image of J is a cyclic group, and is trivial or order 2 except in case ${\displaystyle n=4k+3}$ , in which case it can be large, with its order related to the Bernoulli numbers. The quotient group ${\displaystyle \pi _{n}^{S}/J}$  is the "hard" part of the stable homotopy groups of spheres, and accordingly ${\displaystyle \Theta _{n}/bP_{n+1}}$  is the hard part of the exotic spheres, but almost completely reduces to computing homotopy groups of spheres. The map is either an isomorphism (the image is the whole group), or an injective map with index 2. The latter is the case if and only if there exists an n-dimensional framed manifold with Kervaire invariant 1, which is known as the Kervaire invariant problem. Thus a factor of 2 in the classification of exotic spheres depends on the Kervaire invariant problem.

The Kervaire invariant problem is almost completely solved, with only the case ${\displaystyle n=126}$  remaining open, although Zhouli Xu (in collaboration with Weinan Lin and Guozhen Wang), announced during a seminar at Princeton University, on May 30, 2024, that the final case of dimension 126 has been settled and that there exist manifolds of Kervaire invariant 1 in dimension 126. Previous work of Browder (1969), proved that such manifolds only existed in dimension ${\displaystyle n=2^{j}-2}$ , and Hill, Hopkins & Ravenel (2016), which proved that there were no such manifolds for dimension ${\displaystyle 254=2^{8}-2}$  and above. Manifolds with Kervaire invariant 1 have been constructed in dimension 2, 6, 14, 30. While it is known that there are manifolds of Kervaire invariant 1 in dimension 62, no such manifold has yet been constructed. Similarly for dimension 126.

The order of the group ${\displaystyle \Theta _{n}}$  is given in this table (sequence A001676 in the OEIS) from (Kervaire & Milnor 1963) (except that the entry for ${\displaystyle n=19}$  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 ${\displaystyle \Theta _{n}}$	1	1	1	1	1	1	28	2	8	6	992	1	3	2	16256	2	16	16	523264	24
${\displaystyle bP_{n+1}}$	1	1	1	1	1	1	28	1	2	1	992	1	1	1	8128	1	2	1	261632	1
${\displaystyle \Theta _{n}/bP_{n+1}}$	1	1	1	1	1	1	1	2	2×2	6	1	1	3	2	2	2	2×2×2	8×2	2	24
${\displaystyle \pi _{n}^{S}/J}$	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
index	–	2	–	–	–	2	–	–	–	–	–	–	–	2	–	–	–	–	–	–

Note that for dim ${\displaystyle n=4k-1}$ , then ${\displaystyle \theta _{n}}$  are ${\displaystyle 28=2^{2}(2^{3}-1)}$ , ${\displaystyle 992=2^{5}(2^{5}-1)}$ , ${\displaystyle 16256=2^{7}(2^{7}-1)}$ , and ${\displaystyle 523264=2^{10}(2^{9}-1)}$ . Further entries in this table can be computed from the information above together with the table of stable homotopy groups of spheres.

By computations of stable homotopy groups of spheres, Wang & Xu (2017) proves that the sphere S⁶¹ has a unique smooth structure, and that it is the last odd-dimensional sphere with this property – the only ones are S¹, S³, S⁵, and S⁶¹.

One of the first examples of an exotic sphere found by Milnor (1956, section 3) was the following. Let ${\displaystyle B^{4}}$  be the unit ball in ${\displaystyle \mathbb {R} ^{4}}$ , and let ${\displaystyle S^{3}}$  be its boundary—a 3-sphere which we identify with the group of unit quaternions. Now take two copies of ${\displaystyle B^{4}\times S^{3}}$ , each with boundary ${\displaystyle S^{3}\times S^{3}}$ , and glue them together by identifying ${\displaystyle (a,b)}$  in the first boundary with ${\displaystyle (a,a^{2}ba^{-1})}$  in the second boundary. The resulting manifold has a natural smooth structure and is homeomorphic to ${\displaystyle S^{7}}$ , but is not diffeomorphic to ${\displaystyle S^{7}}$ . 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 ${\displaystyle \mathbb {C} ^{5}}$  satisfying

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

with a small sphere around the origin for ${\displaystyle k=1,2,\ldots ,28}$  gives all 28 possible smooth structures on the oriented 7-sphere. Similar manifolds are called Brieskorn spheres.

Given an (orientation-preserving) diffeomorphism ${\displaystyle f\colon S^{n-1}\to S^{n-1}}$ , gluing the boundaries of two copies of the standard disk ${\displaystyle D^{n}}$  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 ${\displaystyle \Gamma _{n}}$  to be the group of twisted n-spheres (under connect sum), one obtains the exact sequence

${\displaystyle \pi _{0}\operatorname {Diff} ^{+}(D^{n})\to \pi _{0}\operatorname {Diff} ^{+}(S^{n-1})\to \Gamma _{n}\to 0.}$ 

For ${\displaystyle n>5}$ , every exotic n-sphere is diffeomorphic to a twisted sphere, a result proven by Stephen Smale which can be seen as a consequence of the h-cobordism theorem. (In contrast, in the piecewise linear setting the left-most map is onto via radial extension: every piecewise-linear-twisted sphere is standard.) The group ${\displaystyle \Gamma _{n}}$  of twisted spheres is always isomorphic to the group ${\displaystyle \Theta _{n}}$ . The notations are different because it was not known at first that they were the same for ${\displaystyle n=3}$  or 4; for example, the case ${\displaystyle n=3}$  is equivalent to the Poincaré conjecture.

In 1970 Jean Cerf proved the pseudoisotopy theorem which implies that ${\displaystyle \pi _{0}\operatorname {Diff} ^{+}(D^{n})}$  is the trivial group provided ${\displaystyle n\geq 6}$ , and so ${\displaystyle \Gamma _{n}\simeq \pi _{0}\operatorname {Diff} ^{+}(S^{n-1})}$  provided ${\displaystyle n\geq 6}$ .

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 group 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.

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 proposed for exotic 4-spheres are the Cappell–Shaneson spheres (Sylvain Cappell and Julius Shaneson (1976)) and those derived by Gluck twists (Gluck 1962). Gluck twist spheres 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. Many cases over the years were ruled out as possible counterexamples to the smooth 4 dimensional Poincaré conjecture. For example, Cameron Gordon (1976), José Montesinos (1983), Steven P. Plotnick (1984), Gompf (1991), Habiro, Marumoto & Yamada (2000), Selman Akbulut (2010), Gompf (2010), Kim & Yamada (2017).

• Milnor's sphere
• Atlas (topology)
• Clutching construction
• Exotic R⁴
• Cerf theory
• Seven-dimensional space

• Akbulut, Selman (2010), "Cappell–Shaneson homotopy spheres are standard", Annals of Mathematics, 171 (3): 2171–2175, arXiv:0907.0136, doi:10.4007/annals.2010.171.2171, S2CID 754611
• 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, Bibcode:1966PNAS...55.1395B, doi:10.1073/pnas.55.6.1395, MR 0198497, PMC 224331, PMID 16578636
• Brieskorn, Egbert (1966b), "Beispiele zur Differentialtopologie von Singularitäten", Invent. Math., 2 (1): 1–14, Bibcode:1966InMat...2....1B, doi:10.1007/BF01403388, MR 0206972, S2CID 123268657
• 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, MR 0251736
• Cappell, Sylvain E.; Shaneson, Julius L. (1976), "Some new four-manifolds", Annals of Mathematics, 104 (1): 61–72, doi:10.2307/1971056, JSTOR 1971056
• 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, doi:10.4171/qt/5, S2CID 18029746
• 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, MR 0146807
• Hughes, Mark; Kim, Seungwon; Miller, Maggie (2018), Gluck Twists Of S⁴ Are Diffeomorphic to S⁴, arXiv:1804.09169v1
• Gompf, Robert E (1991), "Killing the Akbulut-Kirby 4-sphere, with relevance to the Andres-Curtis and Schoenflies problems", Topology, 30: 123–136, doi:10.1016/0040-9383(91)90036-4
• Gompf, Robert E (2010), "More Cappell-Shaneson spheres are standard", Algebraic & Geometric Topology, 10 (3): 1665–1681, arXiv:0908.1914, doi:10.2140/agt.2010.10.1665, S2CID 16936498
• Gordon, Cameron McA. (1976), "Knots in the 4-sphere", Commentarii Mathematici Helvetici, 51: 585–596, doi:10.1007/BF02568175, S2CID 119479183
• Habiro, Kazuo; Marumoto, Yoshihiko; Yamada, Yuichi (2000), "Gluck surgery and framed links in 4-manifolds", Series on Knots and Everything, 24, World Scientific: 80–93, ISBN 978-9810243401
• Hill, Michael A.; Hopkins, Michael J.; Ravenel, Douglas C. (2016) [First published as arXiv 2009]. "On the non-existence of elements of Kervaire invariant one". Annals of Mathematics. 184 (1): 1–262. arXiv:0908.3724. doi:10.4007/annals.2016.184.1.1. S2CID 13007483.
• Hirzebruch, Friedrich; Mayer, Karl Heinz (1968), O(n)-Mannigfaltigkeiten, exotische Sphären und Singularitäten, Lecture Notes in Mathematics, vol. 57, Berlin-New York: Springer-Verlag, doi:10.1007/BFb0074355, ISBN 978-3-540-04227-3, MR 0229251 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" (PDF). Annals of Mathematics. 77 (3): 504–537. doi:10.2307/1970128. JSTOR 1970128. MR 0148075. – This paper describes the structure of the group of smooth structures on an n-sphere for n > 4. 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.
• Kim, Min Hoon; Yamada, Shohei (2017), Ideal classes and Cappell-Shaneson homotopy 4-spheres, arXiv:1707.03860v1
• Levine, Jerome P. (1985), "Lectures on groups of homotopy spheres", Algebraic and geometric topology, Lecture Notes in Mathematics, vol. 1126, Berlin-New York: Springer-Verlag, pp. 62–95, doi:10.1007/BFb0074439, ISBN 978-3-540-15235-4, MR 8757031
• Milnor, John W. (1956), "On manifolds homeomorphic to the 7-sphere", Annals of Mathematics, 64 (2): 399–405, doi:10.2307/1969983, JSTOR 1969983, MR 0082103, S2CID 18780087
• Milnor, John W. (1959), "Sommes de variétés différentiables et structures différentiables des sphères", Bulletin de la Société Mathématique de France, 87: 439–444, doi:10.24033/bsmf.1538, MR 0117744
• Milnor, John W. (1959b), "Differentiable structures on spheres", American Journal of Mathematics, 81 (4): 962–972, doi:10.2307/2372998, JSTOR 2372998, MR 0110107
• Milnor, John (2000), "Classification of ${\displaystyle (n-1)}$ -connected ${\displaystyle 2n}$ -dimensional manifolds and the discovery of exotic spheres", in Cappell, Sylvain; Ranicki, Andrew; Rosenberg, Jonathan (eds.), Surveys on Surgery Theory: Volume 1, Annals of Mathematics Studies 145, Princeton University Press, pp. 25–30, ISBN 9780691049380, MR 1747528.
• Milnor, John Willard (2009), "Fifty years ago: topology of manifolds in the 50s and 60s" (PDF), in Mrowka, Tomasz S.; Ozsváth, Peter S. (eds.), 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., vol. 15, Providence, R.I.: American Mathematical Society, pp. 9–20, ISBN 978-0-8218-4766-4, MR 2503491
• Milnor, John W. (2011), "Differential topology forty-six years later" (PDF), Notices of the American Mathematical Society, 58 (6): 804–809
• Montesinos, José M. (1983), "On twins in the four-sphere I" (PDF), The Quarterly Journal of Mathematics, 34 (6): 171–199, doi:10.1093/qmath/34.2.171
• Plotnick, Steven P (1984), Gordon, Cameron McA. (ed.), Fibered knots in ${\displaystyle S^{4}}$  – twisted, spinning, rolling, surgery, and branching, American Mathematical Society, Contemporary Mathematics Volume 35, pp. 437–459, ISBN 978-0-8218-5033-6.
• Wang, Guozhen; Xu, Zhouli (2017), "The triviality of the 61-stem in the stable homotopy groups of spheres", Annals of Mathematics, 186 (2): 501–580, arXiv:1601.02184, doi:10.4007/annals.2017.186.2.3, MR 3702672, S2CID 119147703.
• Rudyak, Yuli B. (2001) [1994], "Milnor sphere", Encyclopedia of Mathematics, EMS Press

• Exotic spheres on the Manifold Atlas
• Exotic sphere home page on the home page of Andrew Ranicki. Assorted source material relating to exotic spheres.
• An animation of exotic 7-spheres Video from a presentation by Niles Johnson at the Second Abel conference in honor of John Milnor.
• The Gluck construction on the Manifold Atlas