The univalent Bloch–Landau constant UU, introduced by Landau in 1929, is the largest constant such that the image of every univalent (conformal) map of the unit disk contains a schlicht disk of radius U∣f′(0)∣U|f'(0)|; equivalently, UU is the infimum over normalized univalent functions of the inradius of the image domain. Determining UU exactly is a classical open problem of geometric function theory: the best known bounds, 0.5708858<U≤0.65639370.5708858 < U \le 0.6563937 (Skinner 2009; Carroll–Ortega-Cerdà 2009), have stood since 2009, and no conjectured exact value is established in the literature.

Background

Bloch's theorem (1920s) states that for every function ff analytic in the unit disk D={z:∣z∣<1}\mathbb{D}=\{z:|z|<1\} with f′(0)=1f'(0)=1 the image contains a schlicht disk — a disk on which ff is one-to-one, because ff maps some subdomain of D\mathbb{D} univalently onto it. The classical Bloch constant BB is the largest constant such that every such ff has a schlicht disk of radius at least BB, and the Landau constant LL is the analogous constant for disks centered at the origin; both are also open. In the same 1929 paper Landau introduced a third constant of the family by restricting to functions that are themselves univalent (one-to-one and holomorphic) in D\mathbb{D}. For univalent ff every disk contained in the image is automatically schlicht, so the relevant quantity is the inradius Rf(D)R_f(\mathbb{D}) of the image domain — the supremum of radii of Euclidean disks contained in f(D)f(\mathbb{D}). The univalent Bloch–Landau constant (also called the schlicht Bloch constant; denoted UU by Landau, A\mathfrak{A} by Robinson, and KK in Finch's survey) is U=inf⁡{Rf(D)∣f′(0)∣: f univalent in D, f(0)=0},U=\inf\Big\{\frac{R_f(\mathbb{D})}{|f'(0)|}:\ f \text{ univalent in } \mathbb{D},\ f(0)=0\Big\}, equivalently U=inf⁡f∈SBfU=\inf_{f\in S} B_f where SS is the normalized univalent class (f(0)=0f(0)=0, f′(0)=1f'(0)=1) and BfB_f is the radius of the largest schlicht disk in f(D)f(\mathbb{D}). The four constants satisfy B≤Bℓ≤L≤UB \le B_\ell \le L \le U, where BℓB_\ell is the locally univalent Bloch constant (the same infimum over analytic ff with f′(0)=1f'(0)=1 and f′≠0f'\ne 0 in D\mathbb{D}). The Koebe one-quarter theorem (the image of a normalized univalent function contains the disk of radius 1/41/4 about f(0)f(0)) gives U≥1/4U \ge 1/4; Landau proved U>0.566U>0.566. Lower bounds have crept up through Reich (0.5690.569), Jenkins (0.57050.5705), Toppila (0.57080.5708), Zhang and Jenkins (0.570880.57088), Xiong (0.5708840.570884), to Skinner (0.57088580.5708858), who optimized a pipeline going back to Landau: reduce to the Bloch class {f∈S:∥f∥=1}\{f\in S:\|f\|=1\}, where ∥f∥=sup⁡z∈D(1−∣z∣2)∣f′(z)∣\|f\|=\sup_{z\in\mathbb{D}}(1-|z|^2)|f'(z)| is the Bloch seminorm, and convert pointwise growth estimates — obtained from coefficient inequalities for univalent functions such as the Fekete–Szegö inequality (a sharp bound on ∣a3−λa22∣|a_3-\lambda a_2^2| for the Taylor coefficients of f∈Sf\in S) — into guaranteed omitted-value and hence schlicht-disk bounds, with numerical optimization over auxiliary parameters. These estimates are delicate and numerically sensitive, and the gains have been confined to the fourth through sixth decimals. Upper bounds come from explicit domains. Since UU is an infimum, any simply connected domain D0D_0 of inradius 11 that is the image of D\mathbb{D} under a conformal map ff with f(0)=0f(0)=0 yields U≤1/∣f′(0)∣U \le 1/|f'(0)|. Robinson (1935) obtained U<0.658U<0.658; a proof that extremal domains — domains realizing the infimum — exist was first written down explicitly in his follow-up paper (1936). Goodman (1945) used a disk with radial slit arcs removed to get U<0.65647U<0.65647; Beller and Hummel (1985) refined the same construction to U<0.6564155U<0.6564155; and Carroll and Ortega-Cerdà (2009) obtained the current best U≤0.6563937U\le 0.6563937 by trimming the slit arcs so that the domain is harmonically symmetric in each removed arc with respect to the origin — harmonic symmetry meaning that the Poisson kernels associated with the two prime ends bounded by an internal boundary arc take equal values at the origin — exploiting Fedorov's explicit solution of the Pólya–Chebotarev problem (finding the continuum of minimal logarithmic capacity containing a given finite point set) for four symmetrically placed points, together with conformal welding (constructing a conformal map by gluing copies of the disk along boundary arcs). The theoretical guidance for these constructions is Jenkins' criterion, extended by Carroll: an extremal domain contains a disk of radius exactly UU (an extremal disk), and it is harmonically symmetric in each internal boundary arc that does not touch the extremal disk. These extremality conditions are necessary but not known to be sufficient, and no conjectured exact value of UU is established in the literature.

Problem Statement

Determine the exact value of the univalent Bloch–Landau constant U=inf⁡{Rf(D)∣f′(0)∣: f univalent (one-to-one and holomorphic) in the unit disk D, f(0)=0},U=\inf\Big\{\frac{R_f(\mathbb{D})}{|f'(0)|}:\ f \text{ univalent (one-to-one and holomorphic) in the unit disk } \mathbb{D},\ f(0)=0\Big\}, where Rf(D)R_f(\mathbb{D}) is the inradius of the image domain f(D)f(\mathbb{D}), i.e. the supremum of radii of Euclidean disks contained in f(D)f(\mathbb{D}); equivalently, U=inf⁡f∈SBfU=\inf_{f\in S} B_f where SS is the class of normalized univalent functions (f(0)=0f(0)=0, f′(0)=1f'(0)=1) and BfB_f is the radius of the largest schlicht disk in f(D)f(\mathbb{D}). The problem is posed at its natural full generality: the whole class of univalent functions on the disk, with no restriction on degree, symmetry, or image geometry, and no restriction on the form of the answer. Currently 0.5708858<U≤0.65639370.5708858<U\le 0.6563937 (Skinner; Carroll–Ortega-Cerdà); determining UU means establishing the sharp lower bound Rf(D)≥U ∣f′(0)∣R_f(\mathbb{D})\ge U\,|f'(0)| for every univalent ff (the global lower bound), together with a matching extremal domain — a simply connected domain realized as f(D)f(\mathbb{D}) for a conformal ff with f(0)=0f(0)=0 — whose inradius-to-derivative ratio equals UU (the matching upper bound).

References

  1. Edmund Landau, 'Über die Blochsche Konstante und zwei verwandte Weltkonstanten', Mathematische Zeitschrift 30 (1929) 608–634. DOI: 10.1007/BF01187791, https://doi.org/10.1007/BF01187791
  2. R. M. Robinson, 'The Bloch constant 𝔄 for a schlicht function', Bulletin of the American Mathematical Society 41 (1935) 535–540. DOI: 10.1090/S0002-9904-1935-06138-5, https://doi.org/10.1090/S0002-9904-1935-06138-5
  3. Raphael M. Robinson, 'Bloch functions', Duke Mathematical Journal 2 (1936) 453–459. DOI: 10.1215/s0012-7094-36-00237-5, https://doi.org/10.1215/s0012-7094-36-00237-5
  4. Ruth E. Goodman, 'On the Bloch-Landau constant for schlicht functions', Bulletin of the American Mathematical Society 51 (1945) 234–239. DOI: 10.1090/S0002-9904-1945-08315-3, https://doi.org/10.1090/S0002-9904-1945-08315-3
  5. Edgar Reich, 'On a Bloch-Landau constant', Proceedings of the American Mathematical Society 7 (1956) 75–76. DOI: 10.1090/S0002-9939-1956-0077619-7, https://doi.org/10.1090/S0002-9939-1956-0077619-7
  6. J. A. Jenkins, 'On the schlicht Bloch constant', Journal of Mathematics and Mechanics 10 (1961) 729–734 (no DOI; zbMATH Open: https://zbmath.org/3167938)
  7. S. Toppila, 'A remark on Bloch's constant for schlicht functions', Annales Academiae Scientiarum Fennicae Series A I no. 423 (1968), 4 pp. DOI: 10.5186/aasfm.1969.423, https://doi.org/10.5186/aasfm.1969.423 (zbMATH Open: https://zbmath.org/3267471)
  8. E. Beller and J. A. Hummel, 'On the univalent Bloch constant', Complex Variables Theory and Application 4 (1985) 243–252. DOI: 10.1080/17476938508814109, https://doi.org/10.1080/17476938508814109
  9. S. Zhang, 'On the univalent Bloch constant' (in Chinese), Acta Scientiarum Naturalium Universitatis Pekinensis 25 (1989) 537–539
  10. J. A. Jenkins, 'A criterion associated with the schlicht Bloch constant', Kodai Mathematical Journal 15 (1992) 79–81. DOI: 10.2996/kmj/1138039528, https://doi.org/10.2996/kmj/1138039528
  11. James A. Jenkins, 'On the Schlicht Bloch constant II', Indiana University Mathematics Journal 47 (1998) 1059–1064. DOI: 10.1512/iumj.1998.47.1519, https://doi.org/10.1512/iumj.1998.47.1519
  12. C. Xiong, 'A note on schlicht Bloch constant', Journal of Nanjing Normal University Natural Science Edition 22 (1999) no. 3, 9–10. MR: 1720389, zbMATH Open: https://zbmath.org/1471004
  13. M. Bonk and A. Eremenko, 'Covering Properties of Meromorphic Functions, Negative Curvature and Spherical Geometry', The Annals of Mathematics 152 (2000) 551–592. DOI: 10.2307/2661392, https://doi.org/10.2307/2661392
  14. Tom Carroll, 'An Extension of Jenkin's Condition for Extremal Domains Associated with the Univalent Bloch-Landau Constant', Computational Methods and Function Theory 8 (2008) 159–165. DOI: 10.1007/BF03321679, https://doi.org/10.1007/BF03321679
  15. Tom Carroll and Joaquim Ortega-Cerdà, 'The univalent Bloch-Landau constant, harmonic symmetry and conformal glueing', Journal de Mathématiques Pures et Appliquées 92 (2009) 396–406. DOI: 10.1016/j.matpur.2009.05.008, arXiv:0806.2282, https://arxiv.org/abs/0806.2282
  16. Brian Skinner, 'The univalent Bloch constant problem', Complex Variables and Elliptic Equations 54 (2009) 951–955. DOI: 10.1080/17476930903197199, https://doi.org/10.1080/17476930903197199
  17. Bappaditya Bhowmik and Sambhunath Sen, 'Improved Bloch and Landau constants for meromorphic functions', Canadian Mathematical Bulletin 66 (2023) 1269–1273. DOI: 10.4153/S0008439523000346, https://doi.org/10.4153/S0008439523000346
  18. Steven Finch, 'Errata and Addenda to Mathematical Constants', arXiv:2001.00578 (v4, 2024), https://arxiv.org/abs/2001.00578 (§7.1 states 0.57088586<U≤0.65639370.57088586 < U \le 0.6563937)
  19. Md Firoz Ali and Shaesta Azim, 'Bloch and Landau constants for meromorphic functions', arXiv:2603.05047 (2026), https://arxiv.org/abs/2603.05047