On the curvature bounded sphere problem in ℝ 3 \mathbb{R}^{3}
Hongda Qiu · 2025
Read On the curvature bounded sphere problem in ℝ 3 \mathbb{R}^{3} by Hongda Qiu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-06245.txt.
Introduction
Hongda Qiu settles part of a shape puzzle about spheres with limited bending, in a 2025 paper. Imagine a smooth closed surface in space whose curves never bend more sharply than a unit sphere. A standing question asks how much volume such a surface must enclose.
Qiu proves a clean stepping stone result. If such a surface fits inside an open ball of radius 2, then the region it bounds must contain a whole unit ball. That fact points toward the broader volume claim posed by Burago and Petrunin. An appendix example shows a different side of the problem. The work tightens what geometers know about gently curved shells.
Catalog record
| CREATOR | Hongda Qiu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 12 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-06245
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
On the curvature bounded sphere problem in ℝ 3 \mathbb{R}^{3} Hongda Qiu Abstract We prove that if a topological sphere smoothly embedded into ℝ 3 \mathbb{R}^{3} with normal curvatures absolutely bounded by 1 1 is contained in an open ball of radius 2 2 , then the region it bounds must contain a unit ball. This result suggests a potential direction for a problem formulated by D.Burago and A.Petrunin asking whether a topological sphere smoothly embedded in ℝ 3 \mathbb{R}^{3} with normal curvatures absolutely bounded by 1 1 encloses a volume of at least 4 3 π \frac{4}{3}\pi . The appendix presents an example illustrating an alternative aspect for this problem. 1 Introduction Volume estimates illustrate how the intrinsic curvature of a manifold constrains its embedding or immersion into an ambient space. These estimates are typically studied in broad generality (see, for instance, [ 1 , Chapter 7.1] , [ 2 , Sections 34-35] ). As the exploition deepens, however, some overlooked topics arise to be interesting, even in the context of simple geometric objects. In the early 21th century, Dmitri Burago and Anton Petrunin [ 3 ] formulated a novel problem: Let S be a topological sphere smoothly embedded in \mathbb{R}^{3} with normal curvatures absolutely bounded by 1 . Is the volume enclosed by S at least that of a unit ball, namely \frac{4}{3}\pi ? 1 1 1 The original problem was considered in a three-dimensional non-positively curved space. It was later realized that the Euclidean case implies the Hadamard case. This problem has inspired two directions of discussion—one aiming to prove a positive answer, and the other seeking a counterexample. Regarding the positive direction, one might intuitively attempt to generalize the Pestov–Ionin theorem [ 4 ] 2 2 2 The original reference is in Russian. For an English reference, see, for instance, [ 5 , Section E, Chapter 7, p. 84] that a plane curve with curvature bounded by 1 contains a unit disc, to its three-dimensional analogue. This idea holds when the body enclosed by the surface is convex, but it fails in general: there exists closed surfaces with normal curvatures absolutely bounded by 1 that enclose volume \geq\frac{4}{3}\pi but do not contain a unit ball [ 6 , p. 11] . Moreover, it is known that the maximal radius of a ball that can be contained in any smooth topological sphere with normal curvatures absolutely bounded by 1 is strictly less than 1 [ 2 , Section 30.4, p. 231] [ 7 ] . 3 3 3 This is an English translation by Richard Bishop of the original Russian work by Lagunov and Fet [ 8 ] , though some steps are only sketched. A question related to the positive direction is whether the unit sphere is locally optimal in the sense that any small enough smooth perturbation increases the volume it encloses. In this work, we prove a slightly stronger statement: Theorem 1 . If the surface S , as defined above, is additionally contained by an open ball of radius 2 , then the region it bounds must contain a unit ball. The proof of Theorem 1 is presented in the next Section. It relies on the observation that the body bounded by S is star-shaped ; that is, every point on S can be joined to O by a line segment and this line segment lies in the open region bounded by S [ 2 , Sections 35.1, p. 256] . Consequently, our method fails when the radius of the open ball is increased to 2+\delta for any positive \delta (for instance, consider S as the unit sphere centered at a point located 1+\frac{\delta}{2} from O ). This statement offers some insight into the positive direction for the Burago–Petrunin problem. A potential approach to a positive resolution to the Burago–Petrunin problem is to divide it into two cases: one where S is contained in some open ball of finite radius and another where it is not. These cases can then be treated separately. One might improve our result by increasing the radius of the bounding sphere in the statement, though this needs additional techniques as explained above. For the latter case, a potential first step is to prove the result when the radius is large. As a remark, however, the statemnt does not yield any affirmative conclusion in either the positive or negative direction towards the Burago–Petrunin problem. There is still a hope to find a counterexample, though its construction might be highly nontrivial. For instance, one might construct a counterexample by modifying Lagunov’s Fishbowl [ 9 ] . 4 4 4 For an English reference, see, for instance, [ 5 , Section C, Chapter 11, p. 115] For completeness, an example suggested by Anton Petrunin [ 3 ] is included in Section 3 as an appendix—it is a surface of genus 2 enclosing a volume of \frac{22}{3}\pi-2\pi^{2}+\epsilon\approx 3.3+\epsilon , where the positive constant \epsilon can be arbitrarily small. Here the value 3.3 is less than the volume of a unit ball, \frac{4}{3}\pi\approx 4.2 . Moreover, the Burago-Petrunin problem might also be approached from alternative perspectives, such as by finding a sufficient and necessary condition for S to enclose a volume of \frac{4}{3}\pi . 2 Proof of Theorem 1 Let \Omega be a sphere of radius 2 centered at the origin O and assume that S is contained in the open ball bounded by \Omega . We begin by showing that the body bounded by S is star-shaped. To achieve this goal, we consider the map projecting every point on S radially onto \Omega and show that it is both invertible and short —that is, it does not increase length. Lemma 1 . The radial projection, denoted by \rho:S\to\Omega , is a diffeomorphism. Moreover, it is short. The proof of Lemma 1 relies on the following lemma. Lemma 2 . Let S be the surface as defined above and x any point on S . Denote by \alpha=\alpha(x) the angle between the vector \overrightarrow{Ox} and the normal vector to S at x . Then there exists a positive constant \epsilon such that \epsilon\leq|\overrightarrow{Ox}|\leq 2\cos{\alpha}. (1) Moreover, the origin O lies in the open region enclosed by S . The second part of Inequality 1 is proved by Petrunin [ 10 , Lemma 4.1] . We prove the first part and the final assertion. Proof . We prove the inequality by contradiction. Suppose that S passes through the origin O . Note that the curvatures of geodesics on S are bounded above by 1 . Consider a geodesic on S of length \pi starting from O (such a geodesic exists because closed geodesics on S have length at least 2\pi ). By the Bow Lemma 5 5 5 The original version is proved by Schur [ 11 ] and generalized by Schmidt [ 12 ] . For an English reference, see [ 5 , Theorem 3.19, p. 51] . , the distance between the two ends of this geodesic is no less than that of a unit semicircular arc, which is 2 . It follows that this geodesic intersects \Omega —a contradiction, and the inequality is proved. Now we turn to the final